Comparison of metal release from various metallic biomaterials in vitro

Yoshimitsu Okazaki ^{, , a} and Emiko Gotoh <sup>b

a</sup> Institute of Mechanical Systems Engineering, National Institute of Advanced Industrial Science and Technology, Ecology-oriented Structural Material Group, 2-1 Namiki 1-chome, Tsukuba, Ibaraki 305-8564, Japan
^b National Institute of Technology and Evaluation, 2 Namiki 1-chome, Tsukuba, Ibaraki 305-0044, Japan

Received 15 September 2003;  accepted 28 January 2004.  Available online 11 March 2004.
Biomaterials
Article in Press, Corrected Proof - Note to users

1. Abstract

To investigate the metal release of each base and alloying elements in vitro, SUS316L stainless steel, Co-Cr-Mo casting alloy, commercially pure Ti grade 2, and Ti-6Al-4V, V-free Ti-6Al-7Nb and Ti-15Zr-4Nb-4Ta alloys were immersed in various solutions, namely,
-medium, PBS(−), calf serum, 0.9% NaCl, artificial saliva, 1.2 mass%
-cysteine, 1 mass% lactic acid and 0.01 mass% HCl for 7 d. The difference in the quantity of Co released from the Co-Cr-Mo casting alloy was relatively small in all the solutions. The quantities of Ti released into
-medium, PBS(−), calf serum, 0.9% NaCl and artificial saliva were much lower than those released into 1.2%
-cysteine, 1% lactic acid and 0.01% HCl. The quantity of Fe released from SUS316L stainless steel decreased linearly with increasing pH. On the other hand, the quantity of Ti released from Ti materials increased with decreasing pH, and it markedly attenuated at pHs of approximately 4 and higher. The quantity of Ni released from stainless steel gradually decreased with increasing pH. The quantities of Al released from the Ti-6Al-4V and Ti-6Al-7Nb alloys gradually decreased with increasing pH. A small V release was observed in calf serum, PBS(−), artificial saliva, 1% lactic acid, 1.2%
-cysteine and 0.01% HCl. The quantity of Ti released from the Ti-15Zr-4Nb-4Ta alloy was smaller than those released from the Ti-6Al-4V and Ti-6Al-7Nb alloys in all the solutions. In particular, it was approximately 30% or smaller in 1% lactic acid, 1.2%
-cysteine and 0.01% HCl. The quantity of (Zr+Nb+Ta) released was also considerably lower than that of (Al+Nb) or (Al+V) released. Therefore, the Ti-15Zr-4Nb-4Ta alloy with its low metal release in vitro is considered advantageous for long-term implants.

Author Keywords: Author Keywords: In vitro test; Corrosion; Metal release; Stainless steel; Cobalt-chromium-molybdenum alloy; Titanium alloy; pH

2. Article Outline

1. Introduction

2. Experimental materials and methods

2.1. Alloy specimens

2.2. Static immersion test

2.3. Statistical analysis

3. Experimental results

3.1. Metal release

3.2. Statistical analysis

3.3. Effect of pH on metal release

3.4. Comparison of metal release among Ti alloys

4. Discussion

5. Conclusions

References

3. 1. Introduction

Stainless steel, Co-based alloys and Ti materials are widely used implant materials in clinical practice with each material having its own advantages. Since the metal release from implants is an important subject, numerous studies, including long-term clinical studies, have been conducted on metal release from orthopaedic implants into body fluids (serum, urine, etc.) [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 and 12]. The toxic effects of metals released from prosthetic implants have been reviewed [1]. Implant components fabricated from Co-based alloys have been reported to produce elevated Co, Cr and Ni concentrations in body fluids [3, 4, 5 and 10]. Significantly elevated metal concentrations in body fluids of patients with Co-Cr alloy metal-on-metal bearing hip implants have been reported [4, 7 and 8]. Ti-6Al-4V alloy has been employed for implants and the cytotoxicity of V has become an issue of concern [1, 12, 13, 14, 15 and 16]. Therefore, many studies on metal release from Ti-6Al-4V alloy and on surface treatments that reduce the quantity of V released have also been reported [17, 18 and 19].

Metals from orthopaedic implants are released into surrounding tissue by various mechanisms, including corrosion, wear, and mechanically accelerated electrochemical processes such as stress corrosion, corrosion fatigue and fretting corrosion. This metal release has been associated with clinical implant failure, osteolysis, cutaneous allergic reactions, and remote site accumulation [9 and 11].

The increase in the incidence of allergy, and the necessity for prolonged use require implants having less metal release. In particular, in selecting suitable materials on the basis of usage conditions and durations, the behavior of metal release from each base and alloying element constituting the metallic biomaterials should be examined using various solutions simulated human body fluids. However, there are insufficient research studies that adequately compare metal release from implant materials using various media under the same experimental conditions. In this study, we conducted static immersion tests using various implant materials to obtain quantitative data required for choosing suitable materials according to various clinical usage conditions and durations, and for selecting the optimum test solution to simulate the body environment. Stainless steel, Co-Cr-Mo alloy, and Ti materials were immersed under the same conditions in
-medium, PBS(−), calf serum, 0.9% NaCl, artificial saliva, 1.2 mass%
-cysteine, 1 mass% lactic acid and 0.01 mass% HCl solutions at 37°C for 7 d. Moreover, the metal release of each base and alloying element was compared with pH.

4. 2. Experimental materials and methods

2.1. Alloy specimens

SUS316L stainless steel, which is used for the manufacture of surgical implants in Japan, specified in the Japanese Industrial Standard (JIS) G 4303 was melted by vacuum-induction melting. After soaking at 1200°C for 3 h, the ingot was forged. The billet soaked at 1200°C for 1 h was hot-worked. After maintaining at 1050°C for 30 min, the plate was quenched in water. Finally, the stainless steel plate was solution-treated at 1050°C for 2 min, and then quenched in water. The Co-Cr-Mo alloy specified in ISO 5832-4 was subjected to vacuum-induction melting, and then vacuum-cast by pouring at 1420°C into a mold manufactured by lost wax process. The ingot was homogenized at 1220°C for 4 h. The Co-Cr-Mo alloy as cast (Co-Cr-Mo casting alloy) was used in the immersion test. Commercially pure Ti (CP Ti) grade 2 (ISO 5832-2), the Ti-6Al-4V alloy (ISO 5832-3), the Ti-6Al-7Nb alloy (ISO 5832-11), and the Ti-15Zr-4Nb-4Ta alloy currently specified for surgical implants in JIS T 7401-4 were subjected to vacuum-arc melting. After
(after soaking: 1100°C-3 h for CP Ti, 1150°C-3 h for Ti-6Al-4V and Ti-6Al-7Nb, 1050°C-4 h for Ti-15Zr-4Nb-4Ta) and
-
forging (starting temperature: 850°C for CP Ti, 930°C for Ti-6Al-4V and Ti-6Al-7Nb, 750°C for Ti-15Zr-4Nb-4Ta), the Ti materials except for the Ti-6Al-7Nb alloy were annealed for 2 h at 700°C. The Ti-6Al-7Nb alloy was annealed for 2 h at 740°C. The chemical compositions of stainless steel, the Co-Cr-Mo casting alloy and Ti materials are shown Table 1.

Table 1. Chemical composition (mass%) of materials used

2.2. Static immersion test

The static immersion test was performed in accordance with the currently specified JIS T 0304 standard for metallic biomaterials. Plate specimens (n=25), each 20×40×1 mm³, were cut from each alloy specimen. Immersion tests were conducted at 37°C using various solutions, namely,
-medium (
-modified Eagle's medium, manufactured by Dainippon Pharmaceutical Co. Ltd., containing NaCl, 6.8 g; KCl, 0.4 g; Na₂HPO₄, 1.15 g; NaH₂PO₄ ·H₂O, 0.2 g/l; and trace amounts of amino acids and vitamins, pH=7.4);
-medium containing 10 vol% fetal bovine serum (Gibco BRL, Div. of Life Tech. Inc.) and 7.5% NaHCO₃ solution (1 vol%), phosphate-buffered saline (PBS(−) produced by Nissui Pharmaceutical Co. Ltd., Japan, containing NaCl, 8 g; KCl, 0.2 g; NaH₂PO₄, 0.14 g; and KH₂PO₄, 0.2 g/l, pH=7.2), membrane-filtered calf serum (Gibco BRL, Div. of Life Tech. Inc., pH=6.9), 0.9 mass% NaCl (prepared using guaranteed reagents manufactured by Nacalai Tesque, Inc., Japan, pH=6.6), artificial saliva (manufactured by Teijin Pharma Ltd., Japan, containing NaCl, 0.844 g; KCl, 1.2 g; CaCl₂, 0.146 g; MgCl₂, 0.052 g; K₂PO₄, 0.342 g/l; and small amounts of thickener and benzoic acid, pH=6.2, viscosity: 6 mPa s), 1 mass% lactic acid (prepared with guaranteed reagents manufactured by Nacalai Tesque, Inc., Japan, pH=2.6), 1.2 mass%
-cysteine (pH=2.1)-simulated amino acid, and 0.05 vol% concentrated HCl (0.01 mass% HCl, pH=2.0). The 1.2 mass%
-cysteine solution was prepared from
-cysteine hydrochloride monohydrate (guaranteed reagents produced by Kanto Kagaku Co., Japan). The 0.05 vol% concentrated HCl solution was prepared from ultrahigh-purity hydrochloric acid concentrate (TAMAPURE-AA-10, Tama Chemicals Co., Ltd., Japan). The plate specimens were surface-finished with waterproof emery paper from 120, 240, 400, 600, 800 up to 1000 grit under running water, and then ultrasonically cleaned (three times for 5 min). These specimens were placed in polypropylene bottles and separately sterilized in an autoclave at 121°C for 30 min. The polypropylene bottles used for the immersion tests were carefully cleaned with 5 vol% concentrated HNO₃ solution and ultrapure water (18.3 M
 cm⁻¹) to remove impurities, and thereafter sterilized in an autoclave. The phosphate-buffered saline solution was sterilized in an autoclave. The other solutions, except for the artificial saliva and 0.01% HCl, were sterilized by passing them through a 0.2-
m-pore membrane filter. The artificial saliva was used as received because it contained small amounts of benzoic acid and thickener. Then a 50-ml volume of each solution was poured into the polypropylene bottles each containing a plate specimen. The bottle lids, except for the bottle containing the
-medium, were tightly sealed to maintain an aseptic condition. The bottle containing the
-medium was not tightly sealed to be able to adjust pH. All the bottles were placed inside an incubator in 95% air-5% CO₂ atmosphere at 37°C for 7 d. The plate specimens after the immersion test were carefully repolished with emery paper from 120 up to 1000 grit under running water so that the influence of corrosion is excluded when the specimen is used again.

The concentrations of various metals released into solution were determined in ppb (ng/ml) by inductively coupled plasma-mass spectrometry (ICP-MS, Yokogawa Analytical Systems HP4500 spectrometer, Japan) with an autosampler (CETAC ASX 500) and graphite-furnace atomic absorption spectrometry (GFAAS, Perkin-Elmer Z5100). Co, Cr, Ni, Mo, Ti, Al, V, Zr, Nb and Ta concentrations were analyzed by ICP-MS, and Fe concentration was analyzed by GFAAS. The isotopic mass numbers used were Co, 59; Cr, 53 or 52; Ni, 60 or 58; Mo, 98 or 96; Ti, 49 or 50; Al, 27; V, 51; Zr, 90 or 91; Nb, 93; and Ta, 181, so as to minimize the influence of the matrix. Single-element standard solutions (SPEX CertiPrep, Inc., 1000 
g/ml) were diluted for use as the Co, Cr, Ni, Mo, Ti, Al, V, Zr, Nb and Ta standard solutions. The working curves were established from at least seven plotted points. ICP-MS was performed in a clean room (class 10,000) and the measurement solutions were prepared in a clean bench (class 100). The analytical detection limits under these conditions were all below 0.05 ng/ml. The solutions to be analyzed were stored in polypropylene tubes cleaned with 5 vol% concentrated HNO₃ and ultrapure water.

A solution without a metal specimen was incubated under similar conditions and used for the blank test. The quantity of metal released (
g/cm²) was estimated using the following formula: (amount of solution: 50 ml)×[(metal concentration in each test solution)−(mean metal concentration in blank test with three bottles)]/(surface area of specimen). The quantity of metal released is considered zero at the metal concentration below that of the blank. The mean quantity of metal released and standard deviation were calculated for five bottles.

2.3. Statistical analysis

The quantity of each metal released from each alloy was statistically analyzed by one-way analysis of variance (ANOVA,
=0.05) for analyzing one factor of solution types. The quantities of Cr and Mo released from SUS316L and Co-Cr-Mo casting alloy, and the quantities of Ti released from CP Ti, Ti-6Al-4V, Ti-6Al-7Nb and Ti-15Zr-4Nb-4Ta alloys were statistically analyzed by two-way ANOVA (
=0.05) for analyzing two factors, namely, the alloy type and the solution type.

5. 3. Experimental results

3.1. Metal release

The metal concentrations in the various solutions used in the immersion test without metals (blank test) are shown in Table 2. The quantities of metal released from SUS316L stainless steel into the various solutions are summarized in Fig. 1. The quantities of the base metal and each alloying element released differed depending on the type of solution. The quantities of Fe, Cr, Ni and Mo released were larger in 1.2%
-cysteine, 1% lactic acid and 0.01% HCl than in the
-medium, PBS(−), calf serum, 0.9% NaCl and artificial saliva. Among the
-medium, PBS(−), calf serum, 0.9% NaCl and artificial saliva (pH>6), the quantity of Fe released was relatively high in 0.9% NaCl, PBS(−) and artificial saliva, those of Cr released was relatively high in calf serum, and that of released Ni was high in calf serum and 0.9% NaCl. The quantity of Ni released in
-medium, and that of Cr and Ni released in PBS(−) and artificial saliva were below the detection limit. The quantity of Mo released was very small in PBS(−) and artificial saliva, and below the detection limit in
-medium, calf serum, and 0.9% NaCl. The quantities of metals released from the Co-Cr-Mo casting alloy are shown in Fig. 2. The difference in the quantity of Co released among the solutions was small. Cr was released in a minute quantity into
-medium, calf serum, 0.9% NaCl, 1% lactic acid, 1.2%
-cysteine and 0.01% HCl. The quantity of Mo released was relatively low in all the solutions.

Table 2. Metal concentrations (ng/ml) in various solutions (blank test) without metal plate at 37°C after 1 week

—: Below detection limit.

(12K)

Fig. 1. Quantity of each metal element released from SUS 316L stainless steel into various solutions at 37°C after 1 week: (a) released Fe; and (b) released Cr, Ni and Mo.

(7K)

Fig. 2. Quantity of each metal element released from Co-Cr-Mo casting alloy into various solutions at 37°C after 1 week: (a) released Co; and (b) released Cr and Mo.

The quantities of each metal released from Ti materials are shown in Fig. 3, Fig. 4, Fig. 5 and Fig. 6. The Ti quantities released into
-medium, calf serum and artificial saliva were much lower and below the detection limit in PBS(−) and 0.9% NaCl. The quantity of Ti released from CP Ti was highest in 0.01% HCl. On the other hand, the quantity of Ti released from all the Ti alloys was highest in 1% lactic acid. V release in a minute quantity was observed in PBS(−), calf serum, artificial saliva, 1% lactic acid, 1.2%
-cysteine and 0.01% HCl; released V was below the detection limit in
-medium and 0.9% NaCl. The quantity of Al released from the Ti-6Al-4V or Ti-6Al-7Nb alloy increased in the order of calf serum, 0.9% NaCl, artificial saliva, 1.2%
-cysteine, 0.01% HCl, and 1% lactic acid. Released Al from Ti-6Al-4V alloy was below the detection limit in
-medium; however, a small quantity of Al from Ti-6Al-7Nb alloy was released into the
-medium. The quantity of Nb released from the Ti-6Al-7Nb alloy was the largest in 1% lactic acid, and below the detection limit in
-medium, PBS(−), calf serum and 0.9% NaCl. The quantity of Ti released from the Ti-15Zr-4Nb-4Ta alloy was much smaller than those released from the Ti-6Al-4V and Ti-6Al-7Nb alloys. The quantities of alloying elements released, Zr, Nb, and Ta, were also very low (Fig. 6(b)).

(6K)

Fig. 3. Quantity of Ti released from commercially pure Ti grade 2 into various solutions at 37°C after 1 week.

(8K)

Fig. 4. Quantity of each metal element released from Ti-6Al-4V alloy into various solutions at 37°C after 1 week: (a) released Ti; and (b) released Al and V.

(8K)

Fig. 5. Quantity of each metal element released from Ti-6Al-7Nb alloy into various solutions at 37°C after 1 week: (a) released Ti; and (b) released Al and Nb.

(6K)

Fig. 6. Quantity of each metal element released from Ti-15Zr-4Nb-4Ta into various solutions at 37°C after 1 week: (a) released Ti; and (b) released Zr, Nb and Ta.

3.2. Statistical analysis

The result of one-way ANOVA showed that the solution type had a statistically significant influence on the quantity of each metal element released from each alloy (p<0.001). The result of two-way ANOVA showed that the alloy type, solution type and the interaction of two factors had a statistically significant influence on the quantities of Cr and Mo released from SUS316L or Co-Cr-Mo casting alloy, and the quantities of Ti released from CP Ti, Ti-6Al-4V, Ti-6Al-7Nb and Ti-15Zr-4Nb-4Ta alloys (p<0.001).

3.3. Effect of pH on metal release

The pH effect on the release of each metal is summarized in Fig. 7, Fig. 8 and Fig. 9. In these figures, the results for pHs 3 and 4 were obtained using 0.05% and 0.01% lactic acid solutions. The pH effect on the quantity of base element released from SUS316L stainless steel, Co-Cr-Mo casting alloy, CP Ti and Ti alloys was compared in Fig. 7. The quantity of Fe released from SUS316L stainless steel decreased linearly with increasing pH. The pH effect on quantity of Co released from the Co-Cr-Mo casting alloy was very small. The quantity of Ti released from Ti materials markedly increased with decreasing pH (pH
4), and it markedly attenuated at pHs of approximately 4 and higher.

(12K)

Fig. 7. Effect of pH on quantity of base element released from SUS316L stainless steel, Co-Cr-Mo casting alloy and four Ti, materials in immersion test at 37°C for 1 week: (a) Fe released from SUS 316L stainless steel (
), and released from Co-Cr-Mo casting alloy (); (b) Ti released from commercially pure Ti grade 2 (
); (c) Ti released from Ti-6Al-4V () or Ti-6Al-7Nb (
) alloy; and (d) Ti release from Ti-15Zr-4Nb-4Ta alloy (
).

(7K)

Fig. 8. Effect of pH on quantity of each alloying element released from SUS 316L stainless steel and Co-Cr-Mo casting alloy in immersion test at 37°C for 1 week: (a) Ni released from SUS 316L (
); (b) Cr released from SUS 316L (
) of Co-Cr-Mo (
); and (c) Mo released from SUS 316L (
) or Co-Cr-Mo ().

(11K)

Fig. 9. Effect of pH on quantity of each alloying element released from Ti-6Al-4V, Ti-6Al-7Nb and Ti-15Zr-4Nb-4Ta alloys in immersion test 37°C for 1 week: (a) Al released from Ti-6Al-4V (
) or Ti-6Al-7Nb () alloy; (b) V released from Ti-6Al-4V alloy (
); (c) Zr released from Ti-15Zr-4Nb-4Ta alloy (); (d) Nb released from Ti-6Al-7Nb (
) or Ti-15Zr-4Nb-4Ta (
) alloy; and (e) Ta released from Ti-15Zr-4Nb-4Ta alloy (
).

The quantity of Ni released from SUS316L stainless steel gradually decreased with increasing pH (Fig. 8). Both the quantities of Cr and Mo released from SUS316L stainless steel or the Co-Cr-Mo casting alloy were smaller at a pH of approximately 4 or higher. The quantity of Al released from the Ti-6Al-4V and Ti-6Al-7Nb alloys gradually decreased with increasing pH ( Fig. 9). The release of V ions in minute quantity tended to be enhanced at a pH of about 4 or lower.

3.4. Comparison of metal release among Ti alloys

The ratio of the average quantity of Ti released from the Ti-6Al-7Nb or Ti-15Zr-4Nb-4Ta alloy to that released from the Ti-6Al-4V alloy (Fig. 10) was calculated using the following equation: ratio of released Ti=(average quantity of Ti released from Ti-6Al-7Nb or Ti-15Zr-4Nb-4Ta alloy)/(average quantity of Ti released from Ti-6Al-4V alloy). The ratio of released alloying ELEMENT=(sum of average quantities of each alloying element released from Ti-6Al-7Nb or Ti-15Zr-4Nb-4Ta alloy)/(sum of average quantities of Al and V released from Ti-6Al-4V alloy). The ratio of Ti released from the Ti-15Zr-4Nb-4Ta alloy was smaller than those from the Ti-6Al-4V and Ti-6Al-7Nb alloys in all the solutions. In particular, it was approximately 30% or smaller in 1% lactic acid, 1.2% cysteine and 0.01% HCl.

(25K)

Fig. 10. Ratios of metals released from Ti-6Al-7Nb or Ti-15Zr-4Nb-4Ta alloy to that from Ti-6Al-4V alloy (Control: 1.0): (a) ratio for released Ti; and (b) ratio for sum of released alloying elements.

6. 4. Discussion

The reasons why the quantity of Ni released from SUS 316L in
-medium, PBS(−) and artificial saliva, and why the quantity of Cr released from SUS 316L or Co-Cr-Mo alloy in PBS(−) and artificial saliva became zero are not clarified using only this immersion test. A more detailed analysis of an oxide film formed on each metal surface in each solution seems to be necessary. Perhaps these results might be related to the low metal concentration as shown in the blank test (Table 2).

The toxic effects of metals (Ni, Co, Cr, Ti, Al and V) released from prosthetic implants have been reviewed [1]. Dermatitis related to Ni toxicity has been commonly reported, and numerous animal studies have shown Ni carcinogenicity through inhalation, intravenous and intramuscular parenteral administration of ground Ni complexes. Cr toxicity seems to be closely related to its valence state. Although studies superficially implicate Cr(VI) as the primary toxic agent, it is considered that Cr(III) is the actual agent of toxicity and mutagenicity. Co toxicity is strongly dose related. A recent report has noted an intense histiocytic and plasma cell reaction in the pseudocapsular tissue surrounding Ti-based total joint arthroplasties that have undergone significant wear. Atomic absorption spectrophotometry revealed 56-3700 mg/g of dry tissue (normal: 0 mg/g). The toxic effect of Al is significant considering recent reports of the involvement of Al in Alzheimer's dementia. For V toxicity, although the pentoxide form is from 3 to 5 times more toxic via the pulmonary route than the trioxide form, tissue studies implicate solubility rather than oxidation state as the parameter indicative of toxicity. V is also considered to be an essential element in the body, but may become toxic at excessive levels [17]. The relative growth ratios of murine fibroblast L929 and murine osteoblastic MC3T3-E1 cells decrease at V concentrations of approximately 0.2-0.4 mg/l in the medium [15].

Co and Cr levels (mean±SD) in plasma of six healthy persons without implants are 0.6±0.3 and 0.5±0.1 
g/l, respectively [4]. Co, Cr and Ni concentrations in healthy persons (21 women, 23 men; aged 26-78 years) are 0.05±0.01, 0.06±0.02 and 0.2±0.1 
g/l, respectively [5]. The ranges of trace element composition in serum from patients without implants have been reported to be as follows: Fe, 0.76-1.87 
g/ml; Cr, 0.038-390 ng/ml; Ni, 0.4-220 ng/ml; and Co, 0.011-146 ng/ml [6]. Ti, Al and V concentrations (mean and range) in serum from patients without implants (21 patients; 11 men and 10 women; mean age: 58; range: 44-72 years) are 2.67 (<2.11-7.92), 2.15 (1.09-6.37), and <0.81 ng/ml (below detection limit), respectively [11]. As mentioned above, the range of metal levels reported in literature is considerably wide considering individual differences.

Implants require both biomedical safety and biomechanical compatibility. In terms of biomechanical compatibility, it is necessary that the mechanical properties and corrosion resistance do not degrade over prolonged use. In particular, high corrosion resistance, resulting in low or negligible ion release, is a requirement for metal implants. Corrosion resistance is due to the formation of a protective passive (oxide) film. The stronger the passive film, the better is the corrosion resistance of the implants. The quantity of released metal changes markedly depending on the nature and strength of the metal-oxide bond, structure (vacancies, interstitial elements, degree of ordering), role of alloying element, composition and thickness of oxide films [17, 18, 19, 20, 21, 22 and 23]. The role of alloying element in a passive film formed on SUS316L, Co-Cr-Mo casting alloy and various Ti alloys by the anodic polarization in calf serum at 37°C was examined by X-ray photoelectron spectroscopy (XPS). XPS spectra from stainless steel and Co-Cr-Mo casting alloy show peaks corresponding to Cr₂O₃, Fe oxide, CoO, Mo oxide as well as those of the various metals [20 and 21]. The oxide film formed on Ti alloy mainly consists of TiO, Ti₂O₃ and TiO₂. The TiO₂ film is thinner and stronger than the Cr₂O₃ film [21].

It is interesting that the quantity of each metal obtained in the various solution types generally reflects the variation of pH (Fig. 7, Fig. 8 and Fig. 9). The quantities of Fe and Ni released from Fe-Cr (Fe: 70, Cr: 19, Ni: 9, C: 0.08 mass%) wire into artificial saliva (NaCl: 0.4 g, KCl: 0.4 g, CaCl₂·2H₂O: 0.795 g, NaH₂PO₄·2H₂O: 0.78 g, Na₂S·9H₂O: 0.005 g, urea: 1 g/l, pH=5.1) at 37°C for 1 week are approximately 0.14 and 0.036 
g/cm², respectively. The quantities of Co and Ni released from Co-Cr (Co: 40, Cr: 20, Ni: 15, Fe: 15, Mo: 7, C: 0.15) alloy wire into the artificial saliva are 0.61 and 0.18 
g/cm², respectively [24]. These values are close to those of the curve drawn in Fig. 7 and Fig. 8. The quantity of Ni released from NiTi wire at 37°C from 1 up to 28 d into artificial saliva (NaCl: 0.4 g, KCl: 0.4 g, CaCl₂·2H₂O: 0.795 g, NaH₂PO₄·H₂O: 0.69 g, KSCN: 0.3 g, Na₂S·9H₂O: 0.005 g, and urea: 1 g/l) whose pH was adjusted to 2.5, 3.75, 5.0 and 6.25 by either lactic acid or sodium hydroxide, is fairly large than that released into artificial saliva from SUS316L stainless steel obtained in this immersion test, and it decreases with increasing pH of 2.5-6.25 [25]. The quantity of Ti released from Ti-6Al-4V alloy into bovine serum at 37°C for 7 d is approximately 0.025 
g/cm² [17]. The quantities of Ti released from commercially pure Ti and Ti-6Al-4V alloy into 0.17 
NaCl plus 0.0027 
EDTA (disodium salt) at 37°C at immersion periods of 200 h are about 0.4 and 1.2 
g/cm², respectively [18]. The quantity (mean±SD) of Ti released from commercially pure Ti grade 2 under shaking condition (80 times/min) for 3 weeks into 0.9% NaCl, artificial saliva, which is the same as that used in this study, and 1% lactic acid are zero, 0.28±0.1, and 5.38±2.85 
g/cm², respectively [26]. The quantity of Ti released from Ti films (1 
m thick) is approximately 0.041 
g/cm² at 7 d immersion in 8 m
ethylenediaminetetraacetatic acid in simulated interstitial electrolyte (EDTA/SIE) solution (pH=7.2) in an incubator at 37°C, 10% O₂, 5% CO₂ and 97±3% humidity [27]. These results are close to those obtained in our immersion test ( Figs. 7(b)). Moreover, the quantities of Ti, Al and V released from Ti-6Al-4V alloy in Hanks' solution (pH=7.6), Hanks' solution containing l-leucine (0.1 
) and Hanks' solution saturated 8-hydroxyquinoline are not detectable for 3 up to 12 weeks. However, elevated quantities of Ti, Al and V released were reported in Hanks' solution containing 2% EDTA (disodium salt) and Hanks' solution containing 0.05 
Na-citrate [28]. This elevated quantity of metal released in Hanks' solution containing EDTA is considered to be due to the chelating effect of EDTA. The quantity of Ti released from NiTi wire in pH
3.75 artificial saliva is generally not detectable [25]. The quantity of Ti released from dental-cast commercially pure Ti increases in the presence of fluoride, and low pHs enhance this effect further [29]. At pHs of 3 and lower, passive current density for commercially pure Ti increases markedly with decreasing pH. The change in passive current is also smaller at pHs of 4 and higher [30]. This tendency of electrochemical dissolution of passive films also supports our results. Dissolution kinetics of commercially pure Ti fibers in SIE, SIE plus human serum (serum/SIE), and SIE plus EDTA were examined [27 and 31].

The quantities of metal released from implants into human body fluids in clinical literature (in vivo) were compared with metal releases observed in this immersion test (in vitro). Ni concentrations in blood (0.51±0.17 
g/l, controls without implants: 0.24±0.05 
g/l), plasma (0.26±0.07 
g/l, controls: 0.13±0.04 
g/l) and urine (2.24±0.82 
g/l, controls: 1.14±0.32 
g/l), and Cr level in plasma (0.19±0.08 
g/l, controls: 0.09±0.02 
g/l) are significantly higher in 20 stainless steel Charnley hip arthroplasties after 10-13 years [2]. Ni (median: 0.65; range: 0.1-4.8 
g/l, upper normal value: 0.6 
g/l), Cr (median: 0.75; range: 0.1-2.4 
g/l, upper normal value: 0.5 
g/l), and Co (median: 1.8; range: 0.1-16.4 
g/l, upper normal value: 0.8 
g/l) concentrations in serum after about 2 years are slightly or significantly elevated in Co-Cr-Mo alloy (ASTM F-75; Cr: 27-30, Mo: 5-7, Ni: max. 2.5, C: max. 0.35) hip prosthesis patients (15 patients with firmly fixed; men: 5; women: 10; median age: 64; range: 58-69 years) [3]. Cr and Co concentrations (mean±SD) in plasma of 15 patients (average age: 68; range: 45-80) with Co-Cr-Mo alloy total hip replacement 7-15 years ago are 1.63±0.45 
g/l (control: 0.5±0.1) and 0.72±0.58 (control: 0.6±0.3), respectively [4]. Ni, Cr and Co concentrations (mean±SD) released into serum for 6-120 weeks implantation from knee or hip prostheses (17 men; 11 women; aged 26-78 years) made of porous-coated Co-Cr-Mo alloy (ASTM F-75, Ni<1 mass%) are 0.4±0.1 
g/l (before surgery: 0.3±0.1), 0.10±0.02 (before surgery: 0.06±0.02) and 0.15±0.04 (before surgery: 0.10±0.02), respectively [5]. Preoperative Cr and Co concentrations (mean and range) in serum are 0.06 (0.015-0.16) and 0.16 (0.15-0.41) ng/ml. However, Cr and Co concentrations (mean and range) in serum of 20 patients (9 men and 11 women; mean age: 66; range: 41-77 years) with Co-Cr-Mo alloy total hip arthroplasty at 36 months postimplantation are 0.15 (0.015-0.62) and 0.27 (1.15-0.91) ng/ml, respectively [10]. It has been reported that significant increases in the quantities of Cr and Co released in body fluids (plasma, blood, serum and urine) are further enhanced by wear from metal-on-metal bearing hip replacements made of Co-Cr-Mo alloy [4, 7 and 8].

Commercially pure Ti and Ti alloys are more corrosion-resistant than stainless steel and Co-Cr-Mo alloy as clearly shown by our immersion test. Commercially pure Ti and Ti alloys are used in orthopedic surgery as implants in the forms of wires, nails, plates and screws for fixation and artificial joints. The composition of the Ti oxide film ranges from TiO₂→Ti₂O₃→TiO→TiO₂ [22]. Titanium is a reactive metal, and disruption or damage to the oxide film is repaired immediately in the presence of air or oxidizing media, as would occur in the human body. In applications where there is no oxygen or in reducing media, such as would occur in a crevice, titanium cannot form the passive film and would not be corrosion resistant. Ti, Al and V concentrations (mean and range, or mean±SD) in serum from patients without implants (21 patients; 11 men and 10 women, mean age: 58; range: 38-72 years) are 4.10 (<2.11-7.92), 2.15±0.51, and below 0.81 (detection limit) ng/ml, respectively. However, Ti, Al and V concentrations in serum of patients with loose total hip replacement (21 patients; 11 men and 10 women, mean age: 58; range: 24-78 years) made of Ti-6Al-4V alloy at a mean duration of 44 months (range: 6-60) are 8.08 (<2.11-17.2), 2.16±0.26 and 1.30 (<0.81-1.60) ng/ml, respectively [9]. Preoperative mean Ti and Al concentrations (range) in serum are 1.4 (1.10-3.60) and 1.32±0.23 ng/ml. However, Ti and Al concentrations (mean and range) in serum of 20 patients (10 men and 10 women; mean age: 54; range: 38-67 years) with commercially pure Ti fiber-metal porous-coated modular Ti-6Al-4V alloy femoral stem and Co-Cr-Mo alloy femoral head at 36 months postimplantation are 4.13 (1.10-11.17) and 1.70±0.42 ng/ml, respectively [10]. Ti concentration (mean: 135.57, range: 24.12-716.94 ng/ml) in serum is approximately 50 times greater in patients with failed Ti-6Al-4V alloy total knee replacements (8 patients; men: 5; women: 3; average age: 64 (range: 54-74 years), mean duration of implantation: 57 months (range: 33-92)), and approximately 10 times greater (mean: 35.83 ng/ml, range: 15.72-80.25 ng/ml) in patients with carbon fiber reinforced polyethylene bearing surfaces (13 patients; men: 6; women: 7; average age: 57 (range: 35-68 years), mean duration of implantation: 47 months (range: 36-61)) than in patients having patellar components (mean: 3.15, range: <2.11-6.25 ng/ml, 10 patients; men: 2; women: 8; average age: 63 (range: 55-70 years), mean duration of implantation: 28 months (range: 21-40) made of ultrahigh-molecular-weight polyethylene and Ti-6Al-4V alloy, and the control subjects (mean: 2.67, range: <2.11-7.92 ng/ml, 21 patients; men: 11; women: 10; average aged: 58 (range: 44-72 years)). For Al and V concentrations in serum (3.74 ng/ml (range: 0.83-6.22) for Al, 0.92 ng/ml (range: <0.81-2.64) for V), except for those of patients with failed patellar replacement, no detectable differences were observed [11]. Elevated serum Ti concentrations may indicate accelerated femoral component wear. Giant-cell reaction occurs around polyethylene debris and wear particles in the case of failed Ti-6Al-4V-alloy total hip replacement. The extent of this giant-cell reaction increases as Ti, Al and V concentrations in the dry tissue increase [12]. The Ti alloy disk with an apatite ceramic pin is worn in Eagle's medium, and the wear powder was sterilized in ethanol and added to the culture medium. The growth ratios of L929 or MC3T3-E1 cells cultured with the Ti-6Al-4V alloy wear powder relative to the control cells markedly decreased, and were lower than those cultured with the Ti-15Zr-4Nb-4Ta alloy wear powder. [16]. Corrosion resistance is reduced under the wear condition compared with the static condition in vitro [32, 33 and 34] and in vivo [11 and 12]. Metal release from implanted prostheses is accelerated by biomechanical breakdown of oxide films by fretting wear and micromotion between implanted components and bone by loading or joint function, as well as effects of complex environmental factors [27 and 31]. At pHs of 4 and lower, the quantity of Ti released from Ti materials markedly increased with decreasing pH, and the pH effect on Ti release was markedly attenuated at pHs of approximately 4 and higher ( Fig. 7). As expected from these results, the pH of the Ti surface in vivo may also become less than 4 by the complex action of macrophages.

Due to the perceived safety concern and possible toxic effect, different new alloys have been designed without V, and in some cases, without Al [35, 36, 37, 38, 39, 40 and 41]. Therefore, the use of Ti alloy with as little metal release as possible is desirable for long-term implantation. Zr, Nb and Ta additions to Ti show excellent compatibility in both hard and soft tissues, and inhibited metal release [14]. The effects of the concentrations of various metals on cell viability have been examined using media extracted with metal particles. Since the quantities of the metal ions released into the medium were low (<0.3 ng/ml) for Zr, Ta and Nb particle extractions, the relative growth ratios of the L929 and MC3T3-E1 cells were equal to 1 (noncytotoxic) [42]. A near-beta-type alloy Ti-13Zr-13Nb and beta-type alloys, Ti-12Mo-6Zr-2Fe and Ti-15Mo, have been developed for medical use [35, 36 and 37], and are specified in the ASTM F 1713, ASTM F 1813 and ASTM F 2066 standards, respectively. The addition of 0.2% Pd to commercially pure Ti is widely employed industrially, since it markedly improves corrosion resistance in seawater [43]. Our research group previously reported on the effects of Zr, Nb, Ta and Pd for Ti alloy in terms of corrosion resistance in simulated body fluids, their mechanical properties and cytocompatibility with cultured cells [20, 21, 33, 38, 39, 40 and 41]. Zr, Nb, Ta and Pd are added to Ti alloy because the resultant ZrO₂, Nb₂O₅, Ta₂O₅ and PdO strengthen the TiO₂ passive film that forms on the Ti alloy [20 and 21]. Therefore, Ti-15Zr-4Nb-4Ta alloy containing 0.2% Pd exhibits a much better immersion property than Ti-6Al-4V and Ti-6Al-7Nb alloys ( Fig. 4, Fig. 5 and Fig. 6). The Ti-15Zr-4Nb-4Ta alloy exhibited excellent corrosion resistance under friction [33]. The Ti-15Zr-4Nb-4Ta alloy with its low metal release is considered advantageous as a Ti alloy for long-term implants, since it does not require surface treatment to inhibit Ti release.

7. 5. Conclusions

The quantity of Co released from the Co-Cr-Mo casting alloy was relatively small in all the solutions. The quantities of Ti released into
-medium, PBS(−), calf serum, 0.9% NaCl and artificial saliva were much lower than those released into 1.2%
-cysteine, 1% lactic acid and 0.01% HCl.

The pH effect on the quantity of base element released differed among SUS316L stainless steel, the Co-Cr-Mo casting alloy and Ti alloys. The quantity of Fe released from SUS316L stainless steel decreased linearly with increasing pH. The pH effect on quantity of Co released from the Co-Cr-Mo casting alloy was very small. The pH effect on Ti release was markedly attenuated at pHs of approximately 4 and higher.

The quantity of Ni released from SUS316L stainless steel gradually decreased with increasing pH. The quantities of Cr and Mo released from SUS316L stainless steel or the Co-Cr-Mo casting alloy were smaller at a pH of approximately 4 or higher. The quantity of Al released from the Ti-6Al-4V or Ti-6Al-7Nb alloys gradually decreased with increasing pH. A small quantity of V was released into calf serum, PBS(−), artificial saliva, 1% lactic acid, 1.2%
-cysteine and 0.01% HCl.

The quantity of Ti released from the Ti-15Zr-4Nb-4Ta alloy was smaller than those released from the Ti-6Al-4V and Ti-6Al-7Nb alloys in all the solutions. In particular, it was approximately 30% or smaller in 1% lactic acid, 1.2%
-cysteine and 0.01% HCl. The quantity of (Zr+Nb+Ta) released was considerably lower than that of (Al+Nb) or (Al+V) released. Therefore, Ti-15Zr-4Nb-4Ta alloy with its low metal release is considered advantageous for long-term implants.

8. References

1. K.L. Wapner, Implications of metallic corrosion in total knee arthroplasty. Clin Orthop Relat Res 271 (1991), pp. 12-20. Abstract-MEDLINE | Abstract-EMBASE  

2. U.E. Pazzaglia, C. Minola, L. Ceciliani and C. Riccardi, Metal determination in organic fluids of patients with stainless steel hip arthroplasty. Acta Orthop Scand 54 (1983), pp. 574-579. Abstract-EMBASE | Abstract-MEDLINE  

3. F.F. Hennig, H.J. Raithel, K.H. Schaller and J.R. Dohler, Nickel-, chrom- and cobalt-concentration in human tissue and body fluids of hip prosthesis patients. J Trace Elem Electrolytes Health Dis 6 (1992), pp. 239-243. Abstract-MEDLINE | Abstract-EMBASE  

4. U.E. Pazzaglia, C. Minoia, G. Gualtieri, I. Gualtieri, C. Riccardi and L. Ceciliani, Metal ions in body fluids after arthroplasty. Acta Orthop Scand 57 (1986), pp. 415-418. Abstract-EMBASE | Abstract-MEDLINE  

5. F.W. Sunderman, Jr, S.M. Hopfer, T. Swift, W.N. Rezuke, L. Ziebka, P. Highman, B. Edwards, M. Folcik and H.R. Gossling, Cobalt, chromium, and nickel concentrations in body fluids of patients with porous-coated knee or hip prostheses. J Orthop Res 7 (1989), pp. 307-315. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Compendex  

6. R. Michel, M. Nolte, M. Reich and F. Loer, Systemic effects of implanted prostheses made of cobalt-chromium alloys. Arch Orthop Trauma Surg 110 (1991), pp. 61-74. Abstract-MEDLINE | Abstract-EMBASE  

7. A.W. Schaffer, A. Pilger, C. Engelhardt, K. Zweymueller and H.W. Ruediger, Increased blood cobalt and chromium after total hip replacement. Clin Toxicol 37 (1999), pp. 839-844. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Elsevier BIOBASE   | Full Text via CrossRef

8. W. Brodner, P. Bitzan, V. Meisinger, A. Kaider, F. Gottsauner-Wolf and R. Kotz, Elevated serum cobalt with metal-on-metal articulating surfaces. J Bone Jt Surg 79-B (1997), pp. 316-321. Abstract-EMBASE | Abstract-MEDLINE   | Full Text via CrossRef

9. J.J. Jacobs, A.K. Skipor, J. Black, R.M. Urban and J.O. Galante, Release and excretion of metal in patients who have a total hip-replacement component made of titanium-base alloy. J Bone Jt Surg 73-A (1991), pp. 1475-1486. Abstract-MEDLINE | Abstract-EMBASE  

10. J.J. Jacobs, A.K. Skipor, L.M. Patterson, N.J. Hallab, W.G. Paprosky, J. Black and J.O. Galante, Metal release in patients who have had a primary total hip arthroplasty. J Bone Jt Surg 80-A (1998), pp. 1447-1458. Abstract-EMBASE | Abstract-MEDLINE  

11. J.J. Jacobs, C. Silverton, N.J. Hallab, A.K. Skipor, L. Patterson, J. Black and J.O. Galante, Metal release and excretion from cementless titanium alloy total knee replacements. Clin Orthop Relat Res 358 (1999), pp. 173-180. Abstract-EMBASE | Abstract-MEDLINE   | Full Text via CrossRef

12. H.J. Agins, N.W. Alcock, M. Bansal, E.A. Salvati, P.D. Wilson, Jr., P.M. Pellicci and P.G. Bullough, Metallic wear in failed titanium-alloy total hip replacements. J Bone Jt Surg 70-A (1988), pp. 347-356. Abstract-MEDLINE | Abstract-EMBASE  

13. S.G. Steinemann, Corrosion of surgical implants-in vivo and in vitro tests. In: G.D. Winter, J.L. Leray and K. de Goot, Editors, Evaluation of biomaterials, advances in biomaterials vol. 1, Wiley, Chichester (1980), pp. 1-34.

14. S.G. Steinemann, Compatibility of titanium in soft and hard tissue—the ultimate is osseointegration. In: H. Stallforth and P. Revell, Editors, Materials for medical engineering, Euromat '99 vol. 2, Wiley-VCH, Weinheim (1999), pp. 199-203.

15. Y. Okazaki, S. Rao, S. Asao, T. Tateishi, S. Katsuda and Y. Furuki, Effects of Ti, Al and V concentrations on cell viability. Mater Trans JIM 39 (1998), pp. 1053-1062. Abstract-Compendex  

16. Y. Okazaki and E. Nishimura, Effect of metal released from Ti alloy wear powder on cell viability. Mater Trans JIM 41 (2000), pp. 1247-1255. Abstract-INSPEC | Abstract-Compendex  

17. M. Browne and P.J. Gregson, Surface modification of titanium alloy implants. Biomaterials 15 (1994), pp. 894-898. Abstract

18. A. Wisbey, P.J. Gregson, L.M. Peter and M. Tuke, Effect of surface treatment on the dissolution of titanium-based implant materials. Biomaterials 12 (1991), pp. 470-473. Abstract

19. B.W. Callen, B.F. Lowenberg, S. Lugowski, R.N.S. Sodhi and J.E. Davies, Nitric acid passivation of Ti6Al4V reduces thickness of surface oxide layer and increases trace element release. J Biomed Mater Res 29 (1995), pp. 279-290. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Compendex  

20. Y. Okazaki, K. Kyo, Y. Ito and T. Tateishi, Effects of Mo and Pd on corrosion resistance of V-free titanium alloys for medical implants. Mater Trans JIM 38 (1997), pp. 344-352. Abstract-Compendex | Abstract-INSPEC  

21. Y. Okazaki, T. Tateishi and 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. 78-84. Abstract-Compendex | Abstract-INSPEC  

22. M.A. Imam and A.C. Fraker, Titanium alloys as implant materials. In: S.A. Brown and J.E. Lemons, Editors, Medical applications of titanium and its alloys: the material and biological issues, ASTM STP 1272, American Society for Testing and Materials, Philadelphia (1996), pp. 3-16. Abstract-Compendex  

23. P. Kovacs and J.A. Davidson, Chemical and electrochemical aspects of the biocompatibility of titanium and its alloys. In: S.A. Brown and J.E. Lemons, Editors, Medical applications of titanium and its alloys: the material and biological issues, ASTM STP 1272, American Society for Testing and Materials, Philadelphia (1996), pp. 163-178.

24. N.R. Gjerdet and H. Hero, Metal release from heat-treated orthodontic archwires. Acta Odontol Scand 45 (1987), pp. 409-414. Abstract-MEDLINE | Abstract-EMBASE  

25. H.H. Huang, Y.H. Chiu, T.H. Lee, S.C. Wu, H.W. Yang, K.H. Su and C.C. Hsu, Ion release from NiTi orthodontic wires in artificial saliva with various acidities. Biomaterials 24 (2003), pp. 3585-3592. SummaryPlus | Full Text + Links | PDF (342 K)

26. M. Koike, S. Nakamura and H. Fujii, In vitro assessment of release from titanium by immersion tests. J Jpn Prosthodont Soc 41 (1997), pp. 675-679.

27. K.E. Healy and P. Ducheyne, The mechanisms of passive dissolution of titanium in a model physiological environment. J Biomed Mater Res 26 (1992), pp. 319-338. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Compendex  

28. N. Bruneel and J.A. Helsen, In vitro simulation of biocompatibility of Ti-Al-V. J Biomed Mater Res 22 (1988), pp. 203-214. Abstract-EMBASE | Abstract-MEDLINE  

29. R. Strietzel, A. Hosch, H. Kalbfleisch and D. Buch, In vitro corrosion of titanium. Biomaterials 19 (1998), pp. 1495-1499. Abstract | PDF (308 K)

30. T. Hurlen and W. Wilhelmsen, Passive behavior of titanium. Electrochim Acta 31 (1986), pp. 1139-1146. Abstract

31. K.E. Healy and P. Ducheyne, Passive dissolution of titanium in biological environments. In: S.A. Brown and J.E. Lemons, Editors, Medical applications of titanium and its alloys: the material and biological issues, ASTM STP 1272, American Society for Testing and Materials, Philadelphia (1996), pp. 179-187. Abstract-Compendex  

32. A. Khanm, R. Williams and D.F. Williams, In vitro corrosion and wear of titanium alloys in the biological environment. Biomaterials 17 (1996), pp. 2117-2126.

33. Y. Okazaki, Effect of friction on anodic polarization properties of metallic biomaterials. Biomaterials 23 (2002), pp. 2071-2077. SummaryPlus | Full Text + Links | PDF (217 K)

34. B.J. Smith and P. Ducheyne, Transitional behavior in Ti-6Al-4V fretting corrosion. In: S.A. Brown and J.E. Lemons, Editors, Medical applications of titanium and its alloys: the material and biological issues, ASTM STP 1272, American Society for Testing and Materials, Philadelphia (1996), pp. 219-230. Abstract-Compendex  

35. L.D. Zardiackas, D.W. Mitchell and J.A. Disegi, Characterization of Ti-15Mo beta titanium alloy for orthopaedic implant applications. In: S.A. Brown and J.E. Lemons, Editors, Medical applications of titanium and its alloys: the material and biological issues, ASTM STP 1272, American Society for Testing and Materials, Philadelphia (1996), pp. 60-75.

36. K.K. Wang, L.J. Gustavson and J.H. Dumbleton, Microstructure and properties of a new beta titanium alloy, Ti-12Mo-6Zr-2Fe, developed for surgical implants. In: S.A. Brown and J.E. Lemons, Editors, Medical applications of titanium and its alloys: the material and biological issues, ASTM STP 1272, American Society for Testing and Materials, Philadelphia (1996), pp. 76-87.

37. A.K. Mishra, J.A. Davidson, R.A. Poggie, P. Kovacs and T.J. FitzGerald, Mechanical and tribological properties and biocompatibility of diffusion hardened Ti-13Nb-13Zr—a new titanium alloy for surgical implants. In: S.A. Brown and J.E. Lemons, Editors, Medical applications of titanium and its alloys: the material and biological issues, ASTM STP 1272, American Society for Testing and Materials, Philadelphia (1996), pp. 96-113.

38. A. Ito, Y. Okazaki, T. Tateishi and Y. Ito, In vitro biocompatibility, mechanical properties, and corrosion resistance of Ti-Zr-Nb-Ta-Pd and Ti-Sn-Nb-Ta-Pd alloys. J Biomed Mater Res 29 (1995), pp. 893-900. Abstract-EMBASE  

39. Y. Okazaki, Y. Ito, A. Ito and T. Tateishi, New titanium alloys to be considered for medical implant. In: S.A. Brown and J.E. Lemons, Editors, Medical applications of titanium and its alloys: the materials and biological issues, ASTM STP 1272, American Society for Testing and Materials, Philadelphia (1996), pp. 45-59. Abstract-Compendex  

40. Y. Okazaki, S. Rao, Y. Ito and T. Tateishi, Corrosion resistance, mechanical properties, corrosion fatigue strength and cytocompatibility of new Ti alloys without Al and V. Biomaterials 19 (1998), pp. 1197-1215. Abstract | PDF (1258 K)

41. Y. Okazaki, A new Ti-15Zr-4Nb-4Ta alloy for medical applications. Curr Opin Solid State Mater Sci 5 (2001), pp. 45-53. SummaryPlus | Full Text + Links | PDF (1170 K)

42. Y. Okazaki, S. Rao, S. Asao and T. Tateishi, Effects of metallic concentrations other than Ti, Al and V on cell viability. Mater Trans JIM 39 (1998), pp. 1070-1079. Abstract-Compendex  

43. Titanium data sheet: commercially pure and modified titanium. In: Boyer R, Welsch G, Collings EW editors. Materials properties handbook: titanium alloys. Cambridge, UK: ASM International; 1998. p. 165-260.

Corresponding author

---