Development of Orthopedic Implants with Highly Biocompatible Ti Alloys

Abstract

The material properties of metallic materials used for manufacturing of orthopedic implants are important for understanding the factors affecting the biological, biomechanical, and biochemical performances of orthopedic implants. This chapter will provide the test method for characterizing potential materials for metallic orthopedic device such as artificial joints and osteosynthesis. Particularly, the alloy design and low-cost manufacturing processes of titanium (Ti) metals, cytocompatibility of metals, biocompatibility and corrosion resistance of Ti alloys, and mechanical compatibility of orthopedic implants are summarized. Future trends on both materials and biological evaluation methods are also introduced here. Three-dimensional (3D) layer manufacturing technologies are expected as new technologies for manufacturing, artificial hip joint stems, acetabular cups, and femoral components and tibial trays of artificial knee joints among others. 3D layer manufacturing technologies are also expected for manufacturing porous materials such as acetabular components. It is possible to obtain marketing approval for highly biocompatible implants that are optimized for the skeletal structures and needs of patients by combining 3D layer manufacturing technologies with imaging technologies such as computed tomography (CT).

1. Introduction

Orthopedics is a medical field that deals with the diagnosis and treatment of diseases of the musculoskeletal system including bones, joints, ligaments, and muscles. Patients range from children to the elderly. The organs treated in orthopedics include the spine, spinal cord, peripheral nerves, joints, arms, and legs. The clinical conditions treated by orthopedic specialists include congenital diseases, degenerative diseases, inflammatory diseases, bone and soft tissue tumors, age-related changes, and traumatic conditions such as fractures and dislocations. As we are entering into a super-aging society, the use of orthopedic implants is increasing yearly with the increasing number of patients with fractures caused by aging and osteoporosis, among others. Figure 1 shows orthopedic implants currently in use. There are various implant products, such as bone plates, metaphyseal plates, bone screws, compression hip screws (CHSs), short femoral (γ) nails, intramedullary nails, artificial femoral heads, artificial hip joints, artificial knee joints, artificial shoulder joints, artificial elbow joints, artificial ankle joints, artificial finger joints, and spinal fixation devices. Biochemical and biomechanical compatibilities as well as biological safety are required for orthopedic implant devices. Therefore, metallic orthopedic devices, which are manufactured from various metals having excellent mechanical properties, are widely used in the orthopedic field. To determine biological safety, and biochemical and biomechanical properties, various chemical, biological, and mechanical evaluations are performed for device development and obtaining marketing approval of implant devices. This chapter will provide the evaluation methods for characterizing potential materials for metallic orthopedic devices such as artificial joints and osteosynthesis devices. Recent chemical, biological, and mechanical test results of new biocompatible materials are introduced here, as well as the development of new orthopedic devices. The compatibility of implants with bone geometry and size matching is especially important for elderly patients with fractures around the joints. The risk of damage to tendons and ligaments around joints after implant arthroplasty is likely to increase when there are discrepancies in size and bone geometry.

Figure 1.

Artificial bones and joints currently in use in aging society.

2. Implantable metals and cytocompatibility of metals

2.1 Main metals used in orthopedic implants

Stainless steel, cobalt (Co)-chromium (Cr)-molybdenum (Mo) alloys, commercially pure titanium (C.P. Ti), and Ti alloys are widely used in orthopedic implants. In particular, the use of Ti-6 mass% aluminum (Al)-4 mass% vanadium (V) (Ti-6Al-4V) alloys has been increasing in various orthopedic implant devices. The biological safety and corrosion resistance of stainless steel are improved by increasing the amount of Cr and Mo alloying elements added to the steel. Fatigue strengths are increased to a level equivalent to those of Ti alloys by adding nitrogen (N) and 20% cold working. The fatigue strengths of an industrial Ti material are improved by increasing the amount of trace elements such as oxygen (O) and iron (Fe), whereas the fatigue strength of C.P. Ti grade 4 can be close to those of Ti alloys by 20% cold working. Ti alloys have higher biocompatibility and corrosion resistance than C.P. Ti owing to zirconium (Zr), niobium (Nb), and tantalum (Ta) elements is added. Moreover, the fatigue strengths of materials are substantially improved by changing the conditions of hot forging and heat treatment.

Ti alloys, C.P. Ti, and stainless steel are globally used in various osteosynthesis devices, whereas Co-28Cr-6Mo alloy is widely used in bearing parts of various artificial joints. Artificial hip joint stems are classified into cement hip stems that are fixed to bone using polymethylmethacrylate (PMMA) bone cement and cementless stems that are fixed to bone by osseointegration without PMMA bone cement. Ti alloys with high biocompatibility are used for the cementless hip stem. Cementless stems are coated with bioactive ceramics such as hydroxyapatite (HA) to enhance new bone formation. Co-28Cr-6Mo alloy and high-N stainless steel with high stiffness and strength are popularly used for the cemented hip stem materials.

2.2 Mechanism of cytotoxicity of metal ions

In recent years, the toxicity of Co ions released from metal-on-metal artificial hip joints has attracted attention. Moreover, there have been reports on aseptic lymphocytic vasculitis-associated lesions (ALVALs) associated with Co ions [1]. There are also several clinical reports on the in vivo effect of metal ions released from orthopedic implants. Particularly, the clinical concern about the toxicity of V ions has been reported [1]. When the relative growth ratio is 1 or lower in the cytotoxicity tests, cytotoxicity increases. V and Al ions strongly inhibit the cell growth of both mouse fibroblast L929 and osteoblastic MC3T3-E1 cells compared with Co, Ni, and lead (Pb) ions, causing a marked decrease in relative growth ratio at a concentration of 0.1 ppm or above [1].

Figure 2 shows a schematic illustration of the incorporation of metal ions into cells. Metal ions are incorporated through various ion channels and bind to proteins and amino acids [1]. To clarify the mechanism of cytotoxicity expression, the quantities of V and other metal ions incorporated into mouse fibroblast L929 and osteoblastic MC3T3-E1 cells have been investigated using an inductively coupled plasma-mass spectrometry (ICP-MS) system [1]. Figure 3 shows the relationship between the concentrations of various metals in the cell culture medium and the mean quantity [(femtogram (fg), 10⁻¹⁵ g] of incorporated metal ions. The quantities per cell of metal ions incorporated into the L929 and MC3T3-E1 cells increase with increasing metal concentration in the culture medium, depending on the metal ion type. Lower quantities of gold (Au), Ti, Zr, Nb, Ta, and Cr ions are released into the medium and also incorporated into the cells. Higher quantities of magnesium (Mg), palladium (Pd), Mo, Fe, Ni, Co, Mn, and tungsten (W) ions are incorporated. Moreover, silver (Ag), Pb, V, Cu, and Zn are incorporated at the highest quantities. As shown in Figure 3b, higher quantities of Zn, Pb, Fe, and V ions are incorporated into the MC3T3-E1 cells. The quantity of V ions incorporated into the MC3T3-E1 cells is considerably higher than those of other metal ions. Thus, the cytotoxicity of a metal ion changes with the quantity of the metal ion incorporated into cells. Particularly, V ions are incorporated into cells through xanthine derived from fetal bovine serum. The strong interactions of Mo, Co, and Ni with amino acids have also been clarified by high-performance liquid chromatography (HPLC) [1].

Figure 2.

Schematic illustration of incorporation of metal ions into cells.

Figure 3.

Relationship between metal concentration in medium and quantity of metal per cell. (a) L929 cells and (b) MC3T3-E1 cells.

The effects of lipopolysaccharide (LPS) (positive control), V, and Ni concentrations in a medium on the relative growth ratio of mouse-macrophage-like J774.1 cells have been investigated in Ref. [1]. The relative growth ratio at approximately 0.05 ppm V ion concentration decreases from 1. For V ion concentration, its IC₈₀ (20% inhibitory concentration) is 0.5 ppm and its IC₅₀ is 0.8 ppm. On the other hand, for Ni ion concentration, its IC₈₀ is 1.5 ppm and its IC₅₀ is 3 ppm. The cytotoxicity of V ions for J774.1 cells is approximately 10-fold than that of Ni ions. The rate of increase in the concentration of nitric oxide (NO) released with the activation of J774.1 cells starts to increase at a concentration of V ions 10 times lower than that of Ni ions [1].

Also, an increase in the concentration of cytokines such as tumor necrosis factor-α (TNF-α) and interleukin-6 (IL-6) causes osteoclast differentiation and promotes bone resorption in the orthopedic implant field. Bone resorption is promoted by osteoclasts around artificial joints. Marked increases in TNF-α and IL-6 concentrations in patients with rheumatic disease accelerate bone resorption [1]. The release of TNF-α from J774.1 cells starts at approximately 0.5 ppm V concentration; the concentrations of IL-6 and transforming growth factor-β (TGF-β) markedly increase at a high rate above 1 ppm V concentration. The Ni concentration required to produce cytokines is higher than the V concentration [1].

2.3 Biocompatibility of various metals

The relationship between the cytocompatibility and polarization resistance of various pure metals is summarized in Figure 4a and b [2, 3, 4, 5]. Cytotoxicity of various pure metals is shown in Figure 4c. V ion cytotoxicity is strongly concentration-dependent as shown in Figure 4d. On the other hand, Zr, Nb, and Ta elements exhibit excellent biocompatibility and corrosion resistance, all of which belong to the vital (loose connective vascularized) group with regard to tissue reactions. In the 70 metals in the periodic table of elements, only Ti and Zr elements show excellent cytocompatibility with both bone-derived mouse osteoblast cells and soft-tissue-derived mouse fibroblast cells.

Figure 4.

(a), (b) Relationship between the polarization resistance and biocompatibility of various pure metals and alloys. (c) Cytotoxicity of various pure metals and (d) that at different V ion concentrations.

3. Low-cost manufacturing processes of biocompatible Ti alloys

3.1 Material design of highly biocompatible Ti alloys

The Ti alloys are classified into three types depending on their microstructure: alpha-type (α-) alloys having a hexagonal-close-packed (hcp) structure, beta-type (β-) alloys having a body-centered-cubic (bcc) structure, and alpha-beta-type(α-β-) alloys having a mixed structure comprising α- and β-phases. In the α-β-Ti alloys, Ti-6Al-4V alloy is widely used for various orthopedic implants. The α-β-Ti alloys have higher fatigue strength than that of the β-Ti alloys. Another α-β-Ti alloy, Ti-15mass% Zr-4mass% Nb-(0–4)mass% Ta [Ti-15Zr-4Nb-(0–4)Ta] alloys, is developed in Japan as an excellent biocompatible alloy for long-term orthopedic implant applications and is standardized in JIS T 7401-4 [6].

Zr, Nb, and Ta elements are effective alloying elements of Ti alloys for resulting excellent long-term biocompatibility and corrosion resistance. However, the excessive addition of Ta and Nb to Ti alloys brings a higher manufacturing cost. Therefore, to develop low-cost manufacturing processes, we have investigated the effects of hot-forging and continuous hot-rolling conditions on the hot forgeability, microstructure, biochemical, and biological properties, tensile properties, and fatigue properties of Ti-15Zr-4Nb-(0–4) Ta alloys [7, 8].

3.2 Low-cost manufacturing processes

To develop the biocompatible orthopedic implant devices, Ti-15Zr-4Nb-(0–4) Ta alloys are vacuum-arc melted. The Ti-15Zr-4Nb-(0–4) Ta alloy ingots are homogenized at approximately 1200–1250°C for more than 5 h and beta(β)-forged from the same temperature to forging ratios (cross section before forging/cross section after forging) of more than 3. Then, beta(β)-forgings are conducted to minimize the beta (β, bcc)-phase at 1000 to 1150°C relative to the forging ratio and the size of the billet. Afterward, α-β-forgings at starting temperature of T_β–30°C are conducted to obtain α(hcp)- and β(bcc)-phases by decoupling the fine β-phase. T_β indicates the β-transus temperature (100 vol% β-phase). Finally, 1-m-long×100-mm-square Ti alloy billets are hot-forged by α-β-forging, which are performed using a 1200-ton forging machine under atmospheric conditions. To prevent the edge of the billet from cracking caused by heat loss, the forging time is minimized by adjusting the forging reduction and forging width/speed. The reheating of the ingot and forging are repeated once or twice to optimize the ingot size and microstructure. Between β- and α-β-forgings, the billet surface is ground with a grinder to prevent cracking due to the oxide scale formed on the ingot surface.

A continuous hot-rolling process for Ti alloys is shown in Figure 5. A continuous hot rolling is conducted using 1-m-long×100-mm-square Ti alloy billets. After maintaining them at T_β–60°C for 2 h, the Ti alloy billets are hot-rolled continuously in the α-β-temperature region (below T_β) at a low rolling speed to prevent an increase in the internal temperature of the rolling rod. After α-β-rolling, the annealing of the Ti alloy is generally followed by the removal of internal stress, heat treatment at 700°C for 2 h to optimize the microstructure, and then cooling in air.

Figure 5.

Schematic illustration of continuous hot rolling of Ti-Zr alloy to obtain rod specimens.

Figure 6 shows the room temperature mechanical properties (0.2% proof strength, σ_0.2%PS; ultimate tensile strength, σ_UTS; total elongation, T.E.; and reduction in area, R.A.) of continuously hot-rolled Ti-15Zr-4Nb-4Ta (Ti-Zr A) and Ti-15Zr-4Nb-1Ta (Ti-Zr B) alloy rods. High strength, high hardness, and excellent ductility are obtained in this continuous hot-rolling process for each rod diameter (12 to 50 mm) [8].

Figure 6.

Effects of rod diameter on (a) room temperature mechanical properties (σ_{0.2% PS}, σ_UTS, T.E., and R.A.) and (b) room temperature Vickers hardness (Hv) of hot-rolled Ti-15Zr-4Nb-4Ta (Ti-Zr A) and Ti-15Zr-4Nb-1Ta (Ti-Zr B) alloys.

4. TEM analysis of passive films formed on Ti alloys

Figure 7a–d shows field emission transmission electron microscopy (FE-TEM) images of the passive oxide films formed on the annealed Ti-6Al-4V alloy, Ti-15Zr-4Nb-1Ta alloy, solution-treated high-N stainless steel, and Co-28Cr-6Mo alloy surfaces by anodic polarization up to 0 V vs. SCE (saturated calomel electrode) in 0.9% NaCl at 37°C [9]. To compare the distribution of each metallic element in the passive oxide films formed on metal surfaces, 0 V vs. SCE is selected as the potential in the passive region. As can be seen from the FE-TEM images, the surface of each passive oxide film is protected by carbon, and a passive oxide film is observed on the metal surface. According to the electron diffraction patterns of the Ti-6Al-4V and Ti-15Zr-4Nb-1Ta alloys, the metals have a hexagonal-close-packed (hcp) structure, whereas the oxide films have an amorphous structure. Thin passive oxide films of 3.2 ± 0.2 nm, 4.2 ± 0.2 nm, 2.3 ± 0.1 nm, and 1.7 ± 0.2 nm thicknesses (mean ± standard deviation) are observed on the Ti-6Al-4V alloy, Ti-15Zr-4Nb-1Ta alloy, high-N stainless steel, and Co-28Cr-6Mo alloy surfaces after anodic polarization up to 0 V vs. SCE, respectively [9]. Energy-dispersive X-ray spectrometry (EDX) analysis revealed that the passive oxide film that formed on Ti-15Zr-4Nb-1Ta alloy consists of TiO₂ containing Zr and small quantities of Nb and Ta. The oxide film on Ti-6Al-4V alloy consists of TiO₂ containing Al and a small quantity of V. The oxide film on high-N stainless steel consists of Cr₂O₃ containing Fe and a small quantity of Ni. The oxide film on Co-28Cr-6Mo alloy consists of Cr₂O₃ containing Co and a small quantity of Mo. Figure 8 shows the changes in the concentration (at%) of each element in the oxide films from the oxide/metal interface to the oxide film surface obtained by the EDX analysis of the oxide films. At the oxide/metal interface, the metal concentration is relatively high, and with increasing distance from the oxide/metal interface, the metal concentration decreases and the oxygen (O) concentration increases. The Ti-15Zr-4Nb-1Ta alloy has a higher oxygen concentration in the oxide film than the Ti-6Al-4V alloy. In the oxide films formed on the high-N stainless steel and Co-28Cr-6Mo alloy surfaces, the concentrations of Fe and Co at the oxide/metal interface are high and decrease as the distance from the oxide/metal interface increases. This high oxygen concentration in the oxide films is due to the formation of ZrO₂, Nb₂O₅, and Ta₂O₅ from Zr, Nb, and Ta, respectively.

Figure 7.

FE-TEM images of oxide films formed on (a) Ti-15Zr-4Nb-1Ta, (b) Ti-6Al-4V, (c) Co-28Cr-6Mo alloy, and (d) high-N stainless steel surfaces by anodic polarization up to 0 V vs. SCE in 0.9% NaCl at 37˚C.

Figure 8.

Changes in metal concentrations (at%) in oxide films formed on (a) Ti-15Zr-4Nb-1Ta, (b) Ti-6Al-4V, (c) Co-28Cr-6Mo, and (d) high-N stainless steel from oxide/metal interface to oxide film surface.

The metal surface of an oxide film (electric double layer) is equivalent to a capacitor. The oxide film resistance (R_P) and capacitance (C_CPE, μF) diagrams of implantable metals have been investigated by electrochemical impedance spectroscopy (EIS) [9]. The oxide film resistance (R_P, MΩ) is expressed by R_P = ε_o·ε_r·k_OX /C_CPE, where ε_o and ε_r are the vacuum permittivity (8.854 × 10⁻⁸ μF/cm) and the relative dielectric constant of the oxide film, respectively, and k_OX is the resistivity (MΩ·cm) of the oxide film. R_P and C_CPE are inversely proportional. The Ti-15Zr-4Nb-(0 to 4) Ta alloys have large R_P (maximum: 13 MΩ·cm²) and small C_CPE (minimum: 12 μF·cm⁻²·s^n − 1, n = 0.94) values.

5. Metal ion release in vitro and in vivo

Concerns such as metal sensitivity remain with regard to the long-term exposure to metal ions released from metallic orthopedic implant devices into the body. Seven-day static immersion tests of alloys are conventionally performed to examine the quantities of metal ions released from an alloy in different solutions and their pH dependence [10]. The effects of pH on the quantities of metal ions released from Ti-6Al-4V and Ti-15Zr-4Nb-4Ta alloys, respectively, are reported in Ref. [10]. Bottles of each solution without a specimen are prepared as blanks. The weekly quantity of each metal ion released (μg/cm²) is estimated as (amount of solution: 50 ml) × [(metal concentration in each solution)–(mean metal concentration of three blank bottles)]/(surface area of specimen).

The quantities of metal ions released from Ti-6Al-4V alloy markedly increase with decreasing pH from 4 to 2, although the changes are relatively small above pH 4. In contrast, even a low pH has virtually no effect on the quantity of each metal ion released from Ti-15Zr-4Nb-4Ta alloy, with the exception of Ti; the quantity of released Ti ions starts to increase below pH 3.5, although more gradually than that of Ti-6Al-4V alloy. Thus, 0.9%NaCl containing HCl (0.9%NaCl+HCl) solution adjusted to pH 2 by adding HCl is suitable for accelerated immersion tests.

In metal ion release test in vivo, the metal quantity in a solution containing dissolved bone tissue is measured using ICP-MS (inductively coupled plasma mass spectrometry). The quantities (μg/g) of various metals released into the bone tissue are calculated by dividing the amount (μg) of each metal in the dissolved bone tissue solution by the weight (g) of lyophilized bone tissue [11]. The Ti quantity in the rat tibia tissue-implanted Ti-15Zr-4Nb-4Ta alloy is considerably lower than that in the tibia tissue-implanted Ti-6Al-4V alloy. The Zr, Ta, and Nb quantities are not significantly higher than those of the control sample (without an implant). The Ti quantity in the tibia tissue-implanted Ti-15Zr-4Nb-4Ta alloy is about 40% lower than that in the tibia tissue-implanted Ti-6Al-4V alloy. The total quantity (Zr + Nb + Ta) of Zr, Nb, and Ta is approximately 20% lower than the total quantity (Al + V) of Al and V.

6. Biological safety evaluation under the accelerated extraction condition

6.1 Accelerated extraction condition for biological safety evaluation

Biological safety evaluation tests of two Ti-Zr alloys are performed under the normal extraction condition in accordance with the ISO 10993 series and under the accelerated extraction condition [12]. Ta-free Ti-15Zr-4Nb (Ti-15-4) and Ti-15Zr-4Nb-1Ta (Ti-15-4-1) alloys are used for various biological safety evaluation tests. Plate specimens each with dimensions of 20 mm × 20 mm × 1 mm (thickness) are cut from these Ti-15-4 andTi-15-4-1 alloys for extraction in the biological safety test. The surface of each plate specimen is polished with 1000-grit waterproof emery paper. Rod specimens with a diameter of 1.2 mm and a length of 2.5 mm are used for rat implantation. Each specimen is ultrasonically cleaned in ethanol.

Cell culture medium containing serum is used for extraction in cytotoxicity test, because it supports cellular growth as well as enables the extraction of both nonpolar and polar substances. For normal extraction in main biological safety tests except for cytotoxicity test, plate specimens are extracted in the 0.9%NaCl solution at 121°C for 1 h. On the other hand, the accelerated extraction (0.9%NaCl+HCl) solution adjusted to pH 2 can be prepared in accordance with ISO 16428 as follows [12]. 1 mol/L hydrochloric acid is added to 0.9%NaCl (physiological saline) solution, and the mixed (0.9%NaCl+HCl) solution is adjusted to pH 2. All test specimens are ultrasonically cleaned in ethanol and then sterilized in an autoclave at 121°C for 15 min. After drying, the test specimens are immersed for 7 d in the accelerated (0.9%NaCl+HCl, pH = 2) solution at 37°C. After sufficient stirring, the pH of the accelerated (0.9%NaCl+HCl) solution after extraction is neutralized to 6 ± 1 using 0.1 and 1 mol/L NaOH solutions to obtain various biological evaluation test extracts. Blank extracts (blank control) are similarly prepared without the Ti-15Zr-4Nb-(0 to 4) Ta alloy specimen. Positive control and negative (blank control) groups, and test specimen-treated group are established under different conditions necessary for each biological safety evaluation test to confirm the sensitivity of the test systems.

In implantation tests, new bone formation, bone contact, and osteoid formation rates are compared between test and control specimens. The behavior of newly formed bone tissue around implants can be calculated using the following three parameters. New bone formation rate (%) = (total length of new bone formed around implant)/(length of surrounding implant), bone contact rate (%) = (total length in direct contact with implant)/(length of surrounding implant) × 100, and osteoid formation rate (%) obtained by Villanueva staining = [(total area of osteoid bone)/total area of new bone (osteoid plus calcified bone)].

6.2 Test results of biological safety evaluation

Biological safety evaluation test results obtained with Ti-15Zr-4Nb (Ti-15-4) and Ti-15Zr-4Nb-1Ta (Ti-15-4-1) alloys are reported in Ref. [12]. The effects of normal condition (in 0.9%NaCl and medium) and accelerated condition (in 0.9%NaCl+HCl, pH 2) extracts are compared using Ti-15Zr-4Nb alloys. All of the tests performed in accordance with the ISO 10993 series show no effect (negative) of either extract.

1. No decrease (93–101%) in the colony formation rate is obtained by the colony formation tests at six (3.13, 6.25, 12.5, 25, 50, and 100%) concentrations of the medium extracts for the Ti-15Zr-4Nb and Ti-15Zr-4Nb-1Ta alloys. These results indicate that the Ti-15Zr-4Nb and Ti-15Zr-4Nb-1Ta alloys are noncytotoxic.

2. For skin sensitization evaluation (in maximization tests), an average score of 1 or higher is considered positive for skin reactions in accordance with the ISO 10993-10 standard. The average scores for the accelerated condition (in 0.9%NaCl+HCl) extracts are 0 for the test specimens subjected to 24- and 48-h treatments, indicating that the Ti-15Zr-4Nb and Ti-15Zr-4Nb-1Ta alloys caused no sensitization (no erythema) response.

3. Normal condition and accelerated condition extracts of 0.2 mL each are intradermally injected in the irritation tests of rabbits. No edema and erythema are observed at any of the injection sites and the observation times in all groups, resulting in a score of 0 for intradermal reactions. No significant differences in scores are observed among the blank extract-injected, normal condition extract-injected, and accelerated condition extract-injected groups, indicating that the Ti-15Zr-4Nb alloy causes no intradermal reactions.

4. For the acute systemic toxicity of normal condition and accelerated condition extracts, no effects of the extracts on the general conditions or the weight of any of the mice and no abnormalities are observed in either intrapleural or intraperitoneal organs. This result indicates that the normal condition and accelerated condition extracts from the test Ti-15Zr-4Nb alloy specimens have no acute systemic toxicity.

   For normal condition (in 0.9%NaCl at 121°C for 1 h) extracts intravenously injected to rats for 21 days, there are no significant differences between the control and test Ti-15Zr-4Nb and Ti-15Zr-4Nb-1Ta alloy groups for both female and male rats in the weight, the quantity of food intake, and the results of the urine test. No toxicological significance is observed in the weights of organs and the results of the blood biochemical and hematologic tests. No abnormalities in the macroscopic examination of the systemic organs are also observed for both female and male rats. Moreover, no noteworthy abnormalities are observed in the histopathological examination for both female and male rats. These results show that no clear systemic toxicity is expressed when the normal condition extracts with the Ti-15Zr-4Nb and Ti-15Zr-4Nb-1Ta alloys are intravenously injected to female and male rats once a day for 21 days.

5. In the genotoxicity tests, (a) no increase in the number of revertant colonies is found for gene mutation inducibility. The Ti-15Zr-4Nb alloy immersed in the accelerated extraction 0.9%NaCl+HCl solution shows no gene mutation inducibility. (b) For chromosomal aberration inducibility, no increase in the frequency of appearance of cells with chromosomal aberrations is found. This indicates that the Ti-15Zr-4Nb and Ti-15Zr-4Nb-1Ta alloys do not induce chromosomal aberrations.

6. No cellular infiltration in the implantation tests is observed around the Ti-15Zr-4Nb and Ti-15Zr-4Nb-1Ta alloy specimens, and no degeneration, necrosis, bleeding, or other tissue reactions are found. The formation of new bone is observed around the test specimen in the histopathological examination: The new bone is in direct contact with the Ti-15Zr-4Nb and Ti-15Zr-4Nb-1Ta alloys and is calcified for all rats. Similar reactions are observed for the sites where the control Ti-6Al-4V alloy is implanted. These results indicate that the Ti-15Zr-4Nb and Ti-15Zr-4Nb-1Ta alloys are not inflammatory but osteoconductive, similar to the control Ti-6Al-4V alloy.

7. Fatigue property of Ti alloys

7.1 Fatigue test method

Fatigue resistance is one of the most important mechanical characteristics of structural biomaterials because biomaterials are generally used under cyclic loading conditions. Tension-tension fatigue tests are conducted in accordance with JIS T 0309 [13]. The fatigue tests are conducted in the tension-to-tension mode using sine wave. The testing conditions are a frequency of 10 or 15 Hz and a stress ratio (R = maximum stress/minimum stress) of 10. To obtain S-N curves (profiles of maximum stress vs. number of cycles to failure on logarithmic scale), the test specimens are cycled with a constant load amplitude for a maximum of 10⁸ cycles or until they fail. Hourglass-shaped specimens are popularly used to fracture at the same position of minimum diameter. Rod-shaped specimens exhibit higher fatigue strength than plate-shaped specimens.

7.2 Fatigue strengths of Ti materials

Figure 9 shows the S-N curves obtained by tension-to-tension fatigue tests (10 Hz sine wave) using different types of annealed Ti-15Zr-4Nb-4Ta rod. The arrows on the symbols show the specimens in which fracture does not occur. Figure 9a shows the results obtained with hourglass-shaped and uniform rod specimens, which shows the same tendency. The number of cycles to failure increases with decreasing the maximum cyclic stress. The fatigue strength (maximum cyclic stress) at 1 × 10⁸ cycles is approximately 730 MPa. The effect of specimen diameter on fatigue strength is weak as shown in Figure 9b. The curves of the hourglass-shaped and uniform plate specimens show similar patterns. The fatigue strength of the plate at 1 × 10⁸ cycles is approximately 485 MPa [14]. Higher-grade C.P. Ti shows higher fatigue strength at 1 × 10⁷ cycles: grade 2, 280 MPa; grade 3, 355 MPa; and grade 4, 485 MPa. The fatigue strengths of the annealed Ti-15Zr-4Nb-4Ta, Ti-6Al-4V, V-free Ti-6Al-7Nb, and Ti-6Al-2Nb-1Ta rods are approximately 730, 685, 600, and 700 MPa, respectively. The annealed Ti-6Al-4V alloy consists of an α-phase matrix with an approximately 20 vol% β-phase. The annealed Ti-15Zr-4Nb-4Ta mostly consists of an α-phase matrix with an approximately 25 vol% β-phase. Table 1 shows the tensile properties of various implantable Ti metals. The means and standard deviations of 0.2% proof strength (σ_0.2%PS), ultimate tensile strength (σ_UTS), total elongation (T.E.), and reduction in area (R.A.) are calculated with three test specimens. The fatigue ratios (fatigue strength at 1 × 10⁷ cycles/ultimate tensile strength) are shown in Table 1 and are above 65% for the α-β- and α-type Ti materials.

Figure 9.

S-N curves obtained from tension-to-tension fatigue test with sine wave (10 Hz) in Ringer’s solution at 37°C for annealed Ti-15Zr-4Nb-4Ta rods. (a) Effect of specimen shape (hourglass-shaped and uniform specimens); (b) effect of specimen diameter (3, 4.5, and 6 mm).

Alloy	σ0.2%PS/MPa	σUTS/MPa	T.E. (%)	R.A. (%)	E/GPa	σFS/σUTS
Ti-15Zr-4Nb-4Ta
Annealed (Plate)	800±14	910±10	19±2			0.54
Annealed (Rod)	848±2	915±3	21±2	55±3	94±2	0.80
S.T. +Aged (Rod)	894±5	1020±8	15±2	48±3	98±2	0.89
C.P. Ti Grade 2	276±6	410±4	40±2	60±6	106±2	0.68
C.P. Ti Grade 3	380±2	540±2	32±2	54±2	108±3	0.66
C.P. Ti Grade 4	600±6	701±8	26±2	46±3	118±2	0.69
Ti-6Al-4V	849±1	934±1	16±1	42±3	102±4	0.73
Ti-6Al-7Nb	845±8	960±10	18±2	47±3	108±2	0.63
Ti-6Al-2Nb-1Ta	842±2	900±3	18±3	43±4	110±1	0.81
Ti-15Mo-5Zr-3Al
Solution-annealed	910±10	930±8	19±2	50±2	92±2	0.38
Hot-forged	963±12	988±10	18±2	50±6	102±2	0.71

Table 1.

Tensile properties of implantable metals and ratios of fatigue strength at 10⁷ cycles(σ_FS) to ultimate tensile strength (σ_UTS) shown for comparison among the alloys tested.

E: Young’s modulus.

8. Hot die forging of artificial hip stems

8.1 Hot die forging method

The conditions for rolling Ti-15Zr-4Nb-(0–4) Ta alloy billets (100 mm square) into rods (e.g., 22 and 25 mm in diameter), which are optimal shapes for the high-temperature forging of artificial hip joint stems, are established as described in Section 3.2. With the β-transus temperature (T_β, 850°C) used as a reference, hot rolling is started at a temperature of T_β–50°C. Ti-15Zr-4Nb-(0–4)Ta alloy billets are continuously hot-rolled and shaped into rods with a diameter of 22 or 25 mm. The round bars [wrought Ti-15Zr-4Nb-(0–4) Ta alloy] are annealed at 700°C for 2 h [15].

Figure 10 shows the hot die forging for manufacturing cementless artificial hip stems. Ti-15Zr-4Nb alloy rods with a diameter of 22 or 25 mm are shaped into artificial hip stems by die forging at a high temperature [15]. Molds of three sizes, small size S, medium size M, and large size L are manufactured for forging artificial hip stems with the same shape as the approved product Alloclassic SL artificial hip stems. Two molds are set the upper and lower parts. Considering the forging ratio at positions where burrs are frequently generated, the swaging technique is used to shape a Ti-15Zr-4Nb alloy rod to be processed into a spindle to reduce the quantity of generated burrs. The Ti-15Zr-4Nb alloy rod is shaped into a spindle so that the forging ratio [(cross-sectional area after forging)/(cross-sectional area before forging)] is 1.5–2.0. The spindle-shaped Ti-15Zr-4Nb alloy specimens are continuously introduced into a high-frequency continuous heat treatment furnace, and die forging is started at a temperature of 740 or 780°C (mainly 780°C). The spindle-shaped Ti-15Zr-4Nb alloy specimens are subjected to bending, rough forging, deburring, and finish forging to obtain three sizes (L, M, and S) of artificial hip stem. The oxidized layer formed on the artificial hip stem surface during hot forging is removed by blasting and pickling after annealing at 700°C for 2 h. Thereafter, the surface of artificial hip stems is grit-blasted to have surface roughness (Ra) of approximately 3 to 5 μm using 24-grit Fuji Random WA high-purity Al₂O₃ particles. This is similar to the Ra of the approved product Alloclassic SL artificial hip stems.

Figure 10.

(a) Mold forging of artificial hip stem at high temperature and (b), (c), (d) hot-forged hip stems of sizes L, M, and S, respectively.

8.2 Microstructure of hot-forged artificial hip stems

Figure 11 shows an (a) optical micrograph and (b) SEM image of the transverse (T) section near the center at the 80-mm position from the head of the annealed Ti-15Zr-4Nb hip stem (size S) after hot forging [15]. As shown in Figure 11b, the β-phase (bcc) appears white in the SEM image. In the optical micrograph and SEM image of the annealed Ti-15Zr-4Nb hip stem, the β-phase that precipitated at the grain boundaries of the α-phase (hcp) matrix is found to be produced by hot forging. Figure 11d and e shows TEM images of the T sections of hot-forged artificial hip stem. Figure 11e shows Miller indices and the calculated and measured interplanar distances (d). There is good agreement between both interplanar distance (d) values. The lattice parameters a = b = 0.295 nm and c = 0.468 nm for α (hcp) Ti [ICDD (International Centre for Diffraction Data) No. 044–1294] and a = b = c = 0.331 nm for β (bcc) Ti (ICDD No. 044–1288) are used in the calculation of d values for the α (hcp)-phase and β(bcc)-phase. The β (bcc)-phase is found to precipitate at the grain boundaries of the α (hcp)-phase matrix from the results of electron beam diffraction analysis. The microstructure of the hot-forged Ti-15Zr-4Nb alloy artificial hip stem is finer than that of the Alloclassic Zweymüller SL artificial hip stem (Ti-6Al-7Nb alloy), as shown in Figure 11c. Similar microstructures are obtained for M and L stems.

Figure 11.

(a) Optical micrograph and (b) SEM and (d) TEM images of Ti-15Zr-4Nb stem hot-forged starting at 780˚C; (e) electron beam diffraction pattern of the encircled area in (d); (c) optical micrograph of Alloclassic SL stem.

8.3 Mechanical property of hot-forged artificial hip stems

Miniature mechanical specimens are cut from the hot-forged hip stems and subjected to tensile tests at room temperature and fatigue tests up to 10⁷ cycles. Each of the five uniform rod specimens shown in Figure 12b [rod diameter, 3 mm; gauge length (GL), 15 mm] is cut from the hip stem at the position shown in Figure 12a. The tensile tests at room temperature are conducted at a crosshead speed of 0.5% of the GL/min until the proof stress reaches 0.2%. The test specimen is then pulled at a crosshead speed of 3 mm/min until the specimen fractures. The tensile properties (σ_0.2%PS, σ_UTS, T.E., and R.A.) are measured with five specimens [15].

Figure 12.

(a) Position of miniature specimens cut from hot-forged artificial hip stems; dimensions of specimens used for (b) room temperature tensile, and (c) fatigue tests.

Tension-tension fatigue tests at room temperature under the air are carried out in accordance with JIS T 0309 [13]. Miniature hourglass-shaped rod specimens with 3 mm in minimum diameter and 50 mm in total length, as shown in Figure 12c, cut from hot-forged artificial hip stems at the position shown in Figure 12a. The tension-tension fatigue tests with cylindrical rods are conducted using a sine wave at a stress ratio R [(maximum cyclic stress (σ_max)/(minimum cyclic stress (σ_min)] of 10 and a frequency of 15 Hz in air. To obtain S-N curves [profiles of maximum stress (maximum applied load/area of cross section) vs. the number of cycles to failure on logarithmic scale)], the rod specimens are cycled at various constant maximum cyclic loads until failure or maximum of 10⁸ cycles. The fatigue strength (σ_FS) at 10⁷ cycles is estimated from the S-N curves.

Table 2 shows the tensile properties (n = 5, mean ± standard deviation) of miniature mechanical specimens cut from the Ti-15Zr-4Nb alloy artificial hip stems annealed at 700°C for 2 h after hot forging at 780 or 740°C. The tensile strength of the hot-forged artificial hip stem tends to be higher than that of the 22 or 25 mm (wrought) Ti-15Zr-4Nb alloy rod before hot forging. Also, the tensile strength of the hot-forged artificial hip stem at 780 or 740°C is close to that of the Alloclassic SL (Ti-6Al-7Nb alloy) artificial hip stem [15].

Ti Alloy	σ0.2%PS/MPa	σUTS/MPa	T.E. (%)	R.A. (%)	σFS/MPa	σFS/σUTS
Hot-forged stems
Ti-15Zr-4Nb rod (before forging)	887±5	942±2	20±1	60±1	785±17	0.83
780˚C Forged Ti-15Zr-4Nb stem	919±10	983±9	21±1	58±2	855±14	0.86
740˚C Forged Ti-15Zr-4Nb stem	912±6	979±7	19±2	55±5	840±5	0.85
SL stem (Ti-6-7)	949±23	1034±23	16±1	54±1	805±26	0.78

Table 2.

Tensile properties (σ_0.2%PS, σ_UTS, and T.E., R.A.), fatigue strength at 10⁷ cycles (σ_FS), and fatigue ratio (σ_FS/σ_UTS) of hot-forged Ti-15Zr-4Nb stems.

The fatigue strength of the Ti-15Zr-4Nb alloy hip stem hot-forged at 780°C is ∼855 MPa and slightly higher than that of the artificial hip stem hot-forged at 740°C, which is higher than those of the Alloclassic SL artificial hip stem and wrought Ti-15Zr-4Nb alloy rod. The σ_FS/σ_UTS (0.85) of the hot-forged Ti-15Zr-4Nb alloy is slightly higher than that of the Alloclassic SL artificial hip stem (0.78). Thus, the fatigue strength of the hot-forged Ti-15Zr-4Nb alloy stem is higher than that of the Alloclassic SL artificial hip stem. It is considered that this improvement in the fatigue strength of the hot-forged Ti-15Zr-4Nb alloy hip stem is attributable to its fine microstructure, as shown in Figure 11a.

8.4 Durability of hot-forged artificial hip stems

Tension-tension durability tests of artificial hip stems are carried out in accordance with ISO 7206-4 standard (third edition) [16]. As shown in Figure 13, the artificial hip stem is fixed at angles of 10° (α) in adduction and β (9°) in flection to stem axis, and at the vertical distance(D) from the head center of the artificial hip prosthesis to the upper level of the fixation. The D is set to 80 mm for the hot-forged Ti-15Zr-4Nb alloy artificial hip prosthesis. α (the angle between the load axis and the stem axis) is 10° (in abduction to stem axis), and β (the angle between the line from the center of the head to the tip of the artificial hip femoral stem and the longitudinal sectional stem axis when viewed from the back) is 9° (in flection to stem axis). The durability tests are carried out under tension-tension loading with a sine wave at a load ratio [(minimum cyclic load (P_min)/(maximum cyclic load (P_max)] of 0.1 and a frequency of 3 Hz in air. The durability limits (PD) at 5 × 10⁶ cycles are determined from the P-N curves (profile of maximum load vs. number of cycles to failure on logarithmic scale). The durabilities of artificial hip femoral stems made of Ti-6Al-7Nb and Ti-6Al-4V alloys, which are globally used in clinical settings, are investigated for comparison with that of the Ti-15Zr-4Nb alloy. The cementless total hip femoral stems used are the Alloclassic Zweymüller SL artificial hip stems.

Figure 13.

Fixation method in durability test using artificial hip stem in accordance with ISO 7206-4.

Figure 14 shows the P-N curves of Ti-15Zr-4Nb (Ti-15-4) alloy artificial hip stems (sizes M and S) and Ti-6Al-7Nb (Ti-6-7) alloy Alloclassic Zweymüller SL artificial hip stems. A durability test is conducted for more than 5 million cycles according to ISO 7206-4 standard [15]. The durability limits (PD) at 5 million cycles are 6800 ± 606 N for the size M stem and 3400 ± 495 N for the size S stem. The PDs of the Alloclassic Zweymüller SL artificial hip stem (Ti-6Al-7Nb alloy) are 6400 ± 463 N for the size M stem and 3000 ± 512 N for the size S stem. The Ti-15Zr-4Nb alloy artificial hip femoral stem hot-forged using the forging technology has a durability limit higher than that of the Ti-6Al-7Nb alloy Alloclassic Zweymüller SL artificial hip femoral stem. It fully satisfies the PD (durability limit) at 5 million cycles of 2300 N specified in ISO 7206-4 standard (third edition) [16]. Thus, the artificial hip stem manufactured by hot forging can be used clinically.

Figure 14.

P-N curves of hot-forged Ti-15Zr-4Nb hip stems (sizes S and M) and Alloclassic SL stems (sizes S and M) obtained from the results of compression bending durability tests.

8.5 Stress analysis for durability of artificial hip stem

The stress analysis of the tension-tension durability test results of artificial hip stems is performed using the fatigue strengths shown in Table 2 [15]. Figure 15 shows the stress analysis for the durability test of artificial hip femoral stem according to ISO 7206-4 standard (third edition). Since the hip force F in A–B section is inclined by 9° from the vertical direction, it is resolved into two components, F₁ = F × cos9° and F₂ = F × sin9°. As the neck angle(α) of the artificial hip femoral stem is 131°, the component force F₁ can be resolved into F₃ = F₁ × cos (180° − α + 10°) = F₁ × cos (190° − α) and F₄ = F₁ × sin (190°). Moreover, F₁ can be decomposed into two components (F₅ = F₁ × cos10° and F₆ = F₁ × sin10°); M_x = F₅ × d₁, M_x = F₆ × d₂, and M_y = F₂ × d₂ [15].

Figure 15.

Stress analysis of results of durability tests in accordance with ISO 7206-4.

When the compressive stress is positive, the net axial stress of the cross section (σ_z) generated at position (x, y) on the A–B plane is given by the following:

$${\mathrm{\sigma }}_Z=\frac{F_5}{\mathit{\text{Cross}}-\mathit{\text{sectional area}}}+\frac{M_x}{I_x}y-\frac{M_y}{I_y}x,$$

$${\mathrm{\sigma }}_z=\frac{F_5}{\mathit{bh}}+\frac{F_5{d}_1-F_6{d}_2}{I_x}y-\frac{F_2{d}_2}{I_y}x.$$E1

Here, the inertia moments are I_x = b × h³/12 and I_y = h × b³/12. The shear stresses τ_ZX and τ_ZY generated in the x and y directions, respectively, by the bending moment can be calculated as follows:

$${\mathrm{\tau }}_{\mathit{zx}}=-\frac{\mathit{Qy}{F}_2}{\mathit{bIy}},{\mathrm{\tau }}_{\mathit{zy}}=\frac{\mathit{Qx}{F}_6}{\mathit{hIx}},$$

$${\mathrm{Q}}_x=\frac{b}{2}\left(\frac{h^2}{4}-y^2\right),{\mathrm{Q}}_y=\frac{h}{2}\left(\frac{b^2}{4}-x^2\right).$$E2

Here, fatigue crack is generated from the corners of the stem surface in this durability test of hip stem. Therefore, since Q_X and Q_Y become zero at the material surface (x = b/2, y = −h/2 mm), the shear stresses (τ_zx and τ_zy) become zero.

In addition, torsion moment (T) is given by T = F₂ × d₁. The shear force generated by the torsion moment is given by

$${\mathrm{\tau }}_{\mathit{zx}}=-\frac{T}{2\mathit{Ix}}\mathrm{y},{\mathrm{\tau }}_{\mathit{zy}}=\frac{T}{2\mathit{Iy}}x.$$E3

The absolute values of the shear force generated by the bending moment and torque are used to calculate τ_zx and τ_zy, which are substituted into the following equation to determine the equivalent stress (σ_eq) using the Von Mises criterion:

$${\mathrm{\sigma }}_{\mathit{eq}}={\left[{{\sigma }_z}^2+3({{\tau }_{\mathit{zx}}}^2+{{\tau }_{\mathit{zy}}}^2)\right]}^{\frac{1}{2}}.$$E4

The equivalent stress (σ_eq) can be used to directly compare the fatigue strengths as shown in Table 3. Table 3 shows the maximum equivalent stress (σ_eq) calculated for the hot-forged Ti alloys with Eq. (4). σ_eq is calculated using the durability limits (x = 3.6 mm, y = −5.5 mm; 3400 N for Ti-15Zr-4Nb alloy and 3000 N for Alloclassic SL) of the S stems. The σ_eq values of the Ti-15Zr-4Nb and Alloclassic SL S stems are 871 and 791 MPa, which are close to those (855 and 805 MPa) shown in Table 3, respectively [15]. The σ_eq/σ_FS values of the Ti-15Zr-4Nb and Alloclassic SL S stems are 1.02 and 0.98, respectively, and a good match is obtained. The same can be calculated for a fracture at the neck. Figure 16 shows the changes in σ_eq as a function of maximum cyclic load (N). This analysis is useful for developing artificial hip joints, identifying the worst specimens, and analyzing the durability test results of hip stems.

Specimen	σeq/MPa	x, y/mm	σFS/MPa	σeq/σFS
Hot-forged Ti-15Zr-4Nb	871	(3.6, −5.5)	855	1.02
Alloclassic SL	791	(3.6, −5.5)	805	0.98

Table 3.

Maximum equivalent stress (σ_eq), coordinates (x, y) of the location of σ_eq, σ_FS, and the ratio of maximum equivalent stress to fatigue limit of A–B cross section.

Figure 16.

Changes in maximum equivalent (σ_eq) as a function of maximum cyclic load.

Figure 17 shows the calculation method of the first moment of area (Q_x and Q_y), second moment of area (I_x and I_y), and shear stress (τ_zx and τ_zy) required for the calculation when the cross-sectional shape of hip stem changes. The racetrack-shaped Q_x shown in Figure 17 can be calculated by numerical integration from the y-coordinate of the breaking position to (h + b)/2 (e.g., numerical calculation in a trapezoidal shape) [15].

Figure 17.

Calculation of first moment of area, second moment of area, and shear stress.

8.6 Implantation of grit-blasted Ti-15Zr-4Nb alloy rods in rabbits

The effects of the maximum pullout properties of grit-blasted Ti-15Zr-4Nb (Ti-15-4) and Ti-15Zr-4Nb-4Ta (Ti-15-4-4) alloys have been investigated by implantation in rabbits. Figure 18 shows a schematic illustration of the method of implantation into the femur of rabbits [12]. For comparison, rods made of grit-blasted Ti alloys (Ti-15Zr-4Nb and Ti-6Al-7Nb), shot-blasted Ti-15Zr-4Nb-4Ta alloy, and smooth-surface and machined Ti-15Zr-4Nb alloys are implanted into the distal epiphysis of the femur. After the grit-blasted, shot-blasted, and machined Ti-15Zr-4Nb rods are implanted for 4, 8, 12, 16, and 24 weeks, the pullout test is carried out at a cross-head speed of 0.5 mm/min. Maximum pullout loads are determined from load-displacement curves.

Figure 18.

Schematic illustration of method of implantation into femur of rabbit.

The effect of surface modification is commonly evaluated by a withdrawal test after the implantation test using rabbits. Figure 19 shows the changes in maximum pullout load after implantation into rabbits as a function of implantation period. The maximum pullout loads of the grit-blasted, shot-blasted, and machined Ti-15Zr-4Nb, Ti-15Zr-4Nb-4Ta, Ti-6Al-7Nb (Ti-6-7), and Ti-6Al-4V (Ti-6-4) alloys increase linearly with implantation period. The pullout load of the grit-blasted Ti-15Zr-4Nb alloy rods is higher than that of the shot-blasted ones. The area ratios of residual Al₂O₃ particles of the grit-blasted Ti-15Zr-4Nb and Alloclassic stem surfaces are 9.1 ± 0.4 and 10.4 ± 1.6%, respectively [12].

Figure 19.

Maximum pullout loads after implantation in rabbits.

8.7 Relationship between durability limit and bending strength of osteosynthesis devices

M-N curves (maximum bending moment vs. number of cycles to failure on logarithmic scale) of osteosynthesis devices are measured for the compression bending and four-point bending durability tests [17]. The maximum bending moments (M) are calculated as M = L × P_m for the compression bending durability test and M = h × P_m/2 for the four-point bending durability test, where P_m is the maximum cyclic load (N) in the compression bending durability or four-point bending test. L is lever arm in compression bending tests. h is the distance between the supporting and loading rollers in four-point bending test. The durability limit of various types of osteosynthesis devices linearly increases with increasing bending strength. Figure 20 shows the relationship between the durability limit (at 10⁶ cycles) and the bending strength plotted on a log-log graph. The relationship (durability limit at 10⁶ cycles) = 0.67 × (bending strength) (N·m) (R² = 0.85) is obtained for bone plates, spinal rods, intramedullary nail rods, CHSs, short femoral nails, and epiphyseal plates by regression. This slope of 0.67 is close to the ratio of the fatigue strength to the tensile strength of the raw metals. The relationship for the highly biocompatible Ti-15Zr-4Nb alloy is also linear [17].

Figure 20.

Relationship between durability limit and bending strength of osteosynthesis devices determined by four-point bending and compression bending tests.

Mechanical properties of metallic screws and bone models are used for mechanical testing have been investigated in Ref. [18].

9. Selective laser melting of Ti alloy artificial hip stems

9.1 Selective laser melting conditions

Ti-15Zr-4Nb-4Ta alloy powder is prepared by LPW Technology Ltd. (Cheshire, United Kingdom). Ti-15Zr-4Nb-4Ta and Ti-6Al-4V (EOS GmbH Electro Optical System, Krailling, Germany) powders are prepared by plasma atomization [15]. The Ti-15Zr-4Nb-4Ta and Ti-6Al-4V powders are selective laser-melted in Ar atmosphere using a system comprising an EOS M290 machine (EOS GmbH Electro Optical System, Krailling, Germany), EOSPRINT v. 1.5 and HCS v. 2.4.14 software, and the Ti64 Performance M291 1.10 parameter set. The laser beam power (P) is 280–300 W; the hatch spacing between scan passes (H) is 0.13–0.14 mm; the laser scan speed (V) is fixed from 1200 to 1300 mm/s; powder-deposited layer (stacking) thickness (T) is 0.03 mm; the laser spot focus diameter is 0.1 mm. The volumetric energy density (E) calculated as E = P/(H·T·V) is about 60 J/mm³ [15].

Cylindrical rods with diameter of 9 mm and height of 50 mm built by selective laser melting are cut from the support materials. The building direction of the cylindrical specimens is set to 90° or 0° direction for the base plate using the Ti-15Zr-4Nb-4Ta powders. The Ti-15Zr-4Nb alloy cylindrical rods after selective laser melting are heat-treated at 760°C for 4 h followed by air cooling. For comparison, Ti-6Al-4V alloy stems and Ti-6Al-4V alloy rods are similarly selective laser-melted. The selective laser-melted Ti-6Al-4V alloy specimens are annealed at 840°C for 4 h followed by air cooling [15].

9.2 Microstructure and mechanical properties of selective laser-melted Ti alloys

Figure 21 shows optical micrographs and SEM and TEM images of the T section of annealed Ti-15Zr-4Nb-4Ta rods after selective laser melting (90° direction) [15]. The selective laser-melted Ti-15Zr-4Nb-4Ta rod has an acicular structure. TEM images of the selective laser-melted Ti-15Zr-4Nb-4Ta rod show that the rods consisted of fine α’ (hcp) lath martensitic structure (lattice parameters a = b = 0.295, c = 0.468 nm) that precipitated with the fine β(bcc)-phase (lattice parameters a = b = 0.331 nm) at the grain boundary of the α’ lath martensitic matrix caused by rapid solidification.

Figure 21.

(a) Optical micrograph and (b) SEM and (c), (d) TEM images of selective laser-melted Ti-15Zr-4Nb-4Ta; (e) electron beam diffraction pattern obtained at the location indicated by P in (d) (precipitation).

Tensile properties of the selective laser-melted Ti-15Zr-4Nb-4Ta and Ti-6Al-4V alloy rods are shown in Table 4. The tensile properties of the selective laser-melted Ti-15Zr-4Nb-4Ta and Ti-6Al-4V rods are similar to those of the hot-forged Ti-15Zr-4Nb rod shown in Table 2. The tensile strengths of the selective laser-melted Ti rods are close to that of the wrought Ti-15Zr-4Nb alloy rod. The tensile properties of the selective laser-melted Ti-15Zr-4Nb-4Ta and Ti-6Al-4V alloy rods fully satisfy those (σ_0.2%PS ≥ 780, σ_UTS ≥ 860 MPa, and T.E. ≥10%) specified in JIS T 7401-4 and ISO 5832-3 [19]. The fatigue strengths (σ_FS) of the 90°-direction-built Ti-15Zr-4Nb-4Ta (once-melted) and Ti-6Al-4V (once- and 10-times-melted) alloy rods are ∼640, ∼680, and ∼ 660 MPa, respectively. It is considered that to increase the fatigue strength of selective laser-melted Ti alloys, it is effective to improve the morphology of the α’ lath martensitic structure and change the α’ lath martensitic structure to α (hcp)-β (bcc)-two-phase structure by heat and hot isostatic press (HIP) treatments. Selective laser melting is a promising new manufacturing technology for artificial hip stems in the future.

Ti/Alloy	σ0.2%PS/MPa	σUTS/MPa	T.E. (%)	R.A. (%)	σFS/MPa	σFS/σUTS
Selective laser-melted rods
Once-melted 0º Ti-15Zr-4Nb-4Ta	880±2	1032±1	14±1	31±2
Once-melted 90º Ti-15Zr-4Nb-4Ta	860±3	1022±2	16±1	36±7	640±11	0.63
Once-melted 90º Ti-6Al-4V	949±3	1041±2	15±1	46±2	680±37	0.65
10-times-melted 90º Ti-6Al-4V	946±2	1036±2	15±1	47±1	660±14	0.64

Table 4.

Tensile properties (σ_0.2%PS, σ_UTS, T.E., and R.A.), fatigue strength at 10⁷ cycles (σ_FS), and fatigue ratio (σ_FS/σ_UTS) of selective laser-melted Ti-15Zr-4Nb-4Ta and Ti-6Al-4V rods.

Dimples are observed on the fracture surfaces after room temperature tensile test. A fatigue crack develops with a fatigue fracture from the internal parts of the specimen, and striations are observed.

9.3 Durability of selective laser-melted artificial hip stems

Tension-tension durability tests of selective laser-melted Ti-15Zr-4Nb-4Ta and Ti-6Al-4V alloy artificial hip stems are performed to obtain the P-N curves (maximum cyclic load vs. the number of cycles to failure on logarithmic scale) according to ISO 7206-4 standard (third edition) [16]. The artificial hip stem is fixed at D (vertical distance from head center of artificial hip prosthesis to upper level of the fixation) = 80 mm and at angles of α (in adduction) = 10° and β (in flection) = 9°, as shown in Figure 13. On the other hand, D was 0.4 × CT in ISO 7206-4 second edition. CT (stem length) is the distance between the head center of the artificial hip prosthesis (C) and the tip of the hip stem (T). At this D = 80 mm fixation, it is assumed that the cement is loose in a cement-type stem. The cementless-type stem, for example Alloclassic SL artificial hip femoral stem, is fixed by the fixing force of the autologous bone from the proximal portion to the distal portion. Therefore, for the selective laser-melted Ti-15Zr-4Nb-4Ta and Ti-6Al-4V alloy artificial stems, considering that the stem is the cementless type, D = 0.4 × CT; α and β are 10° and 9°, respectively. The tension-tension durability tests are performed at room temperature in air with a sine wave at a minimum-to-maximum load ratio of 0.1. The frequency is 3 Hz. The durability limits (PD) at 5 × 10⁶ cycles are estimated from the P-N curves.

Figure 22 shows the L-N curves of the selective laser-melted Ti-15Zr-4Nb-4Ta and Ti-6Al-4V stems and the approved product HA–TCP and S–ROM stems. The durability limit of the selective laser-melted Ti-15Zr-4Nb-4Ta stem is lower because the selective laser melting conditions developed for the Ti-6Al-4V alloy are used and the selective laser melting conditions for the Ti-15Zr-4Nb-4Ta alloy are as yet not developed. On the other hand, the durability limit of the selective laser-melted Ti-6Al-4V stem is ∼2500, which is much higher than those of the approved product HA–TCP and S–ROM stems. With a load of 2300 N and the laser-melted Ti-6Al-4V artificial hip femoral stem fixed at D = 80 mm, the stems brake after around 100,000 cycles. To clarify the durability limit of 2300 N at fixation of 80 mm, it is necessary to consider the selective laser melting conditions of the Ti-6Al-4V hip stem. The durability of artificial hip femoral stem can be improved by examining the annealing conditions after the selective laser melting. On the other hand, in the cementless type, the stem is fixed by the fixing force of the autologous bone from the proximal part to the distal part. Even in the durability test with fixation D = 0.4 CT (52 mm), the fixation position is clinically lower. Therefore, selective laser melting (SLM) can be applied to manufacture of custom-made artificial hip stems [20]. SLM is also a new promising manufacturing technology for fabricating Co-Cr-Mo alloy TKA (Total Knee Arthroplasty) femoral component [21].

Figure 22.

L–N curves of selective laser-melted Ti-15Zr-4Nb-4Ta and Ti-6Al-4V hip stems (size S), approved product HA-TCP fiber metal, and S-ROM hip stems obtained by compression bending durability tests.

10. Evaluation of bearing parts of artificial prostheses

Artificial prostheses such as total hip arthroplasty (THA) and total knee replacement (TKR) are performed in increasing numbers of elderly patients. The differences between bipolar hip arthroplasty (BHA) and total hip arthroplasty (THA) are shown Figure 23. The human hip joint consists of the femur and pelvis. The femoral head fits into the concave part of the pelvis called the acetabulum and is moved by ligaments and muscles around it. The surfaces of the femoral head and acetabulum are covered by an elastic tissue called the cartilage, which absorbs shocks so that the joint moves smoothly and experiences less abrasion. In BHA, only the femoral head is removed and replaced with an artificial femoral head in cases of fractures caused by a fall, for which the preservation of the femoral head is difficult. The functions of the acetabulum, namely, the sliding of the human hip joint, are maintained. On the other hand, in THA, the acetabulum of the pelvis is removed and the entire hip joint is replaced with an artificial hip joint. In contrast to BHA, the evaluation of wear characteristics of the sliding part of the hip joint is required in THA because the acetabular component is secured to the pelvis with bone screws. Most bearing part of the artificial hip joints consists of the acetabular and femoral components. The acetabular component consists of a liner made of ultrahigh-molecular-weight polyethylene (UHMWPE) with high density and average molecular weight and an acetabular cup. The femoral component consists of a stem and a stem femoral head. GUR1020, GUR1050, and GUR 4150 UHMWPE materials are popularly used for the bearing inserts of artificial hip joints.

Figure 23.

Differences between bipolar hip and total hip arthroplasties.

Among the greatest advantages of implant arthroplasty are the recovery from the bedridden state and joint pain relief. The sliding part of an artificial hip joint is generally made of a combination of metal and UHMWPE (metal on polyethylene). As the hard-on-soft bearing of THA, metal-on-polyethylene and ceramic-on-polyethylene bearings are being used worldwide. There are currently three types of UHMWPE material: conventional UHMWPE (CPE), highly cross-linked UHMWPE (XLPE), and vitamin E-containing highly crosslinked UHMWPE (VEPE). Hip simulator wear tests are specified in ISO 14242-1 standard [22]. The volumetric wear rate per million cycles (mm³/Mc) of artificial hip joints made of XLPE markedly decreases when the joints are subjected to a radiation dose of 40 kGy or higher [23]. As shown in Figure 24, the hip volumetric wear rates per million cycles (mm³/Mc) of both CPE and XLPE slightly increase with increasing femoral head diameter [23].

Figure 24.

Effect of femoral head diameter on hip volumetric wear rates of conventional UHMWPE and highly cross-linked UHMWPE.

The penetration of the femoral head into a UHMWPE acetabular cup is evaluated using Eq. (5) [24, 25].

$${\delta }_V={\mathrm{\pi R}}_{1^2}\times \mathrm{p}/\left[1+\left({\mathrm{R}}_2–{\mathrm{R}}_1\right)/\mathrm{p}\right]$$E5

Here, δ_V (volumetric wear) represents the volume of the UHMWPE removed by wear (mm³), p is the penetration (linear wear) of the femoral head into the UHMWPE acetabular cup (mm/year), and R₁ and R₂ are the radii of the femoral head and acetabular cup (mm), respectively. A linear wear of 0.1 mm/year or less is recommended so that osteolysis does not occur; this value is the osteolysis threshold. The UHMWPE wear rate in clinical use can be estimated using Eq. (5). A bearing with a highly cross-linked UHMWPE (XLPE) insert shows a low wear rate, (5 mm³/Mc) suggesting its excellent long-term clinical performance.

In artificial knee prostheses, there is a variation in the constraint of the tibial insert depending on whether the posterior cruciate ligament is preserved [cruciate retaining (CR)] or sacrificed [posterior stabilized (PS)] at the time of knee joint surgery. The wear rates of artificial knee joint UHMWPE inserts are examined using a knee joint simulator that satisfies ISO 14243-1 (load control) [26] or ISO 14243-3 (displacement control) [27] standard. The volumetric wear rate of XLPE knee joint markedly decreases when the joints are exposed to a radiation dose of 40 kGy or higher [21]. Volumetric wear rates of 3 mm³/Mc for XLPE knee joint inserts and 15 mm³/Mc for CPE knee joint inserts are recommended as goals for the development of new knee joints. The volumetric wear rate (7.2±2 mm³/Mc) of the CPE insert against the selective laser-melted (SLM) Co-28Cr-6Mo femoral component is lower than that (15.1±1.2 mm³/Mc) of CPE insert against cast Co-28Cr-6Mo femoral component (Figure 25) [21]. Since the effect of precipitates of the pi (π)-phase [(Cr, Mo, W)₁₂Co₈(C, N)₄] on the increase in the wear rate is negligible, SLM is a promising new manufacturing technology for knee femoral components.

Figure 25.

(a) Selective-laser-melted (as-built) Co-28Cr-6Mo femoral component and (b) cylindrical specimen.

In bipolar hip arthroplasty (BHA), clinically, the bearing part is not excised and the bearing part of the femur is preserved, so it is not necessary to evaluate their wear property using a simulator. In particular, the pullout properties of a bipolar cup are important [28], particularly, the negative correlation between the maximum pullout load and oscillation angle (OA) in the bipolar cup.

11. Future trends

11.1 Orthopedic implant application of highly biocompatible Ti-Zr-Nb alloys

Ti-15Zr-4Nb alloy has excellent room temperature strength and fatigue properties with high biocompatibility. Moreover, for Ti-Zr-Nb alloy, a body-centered cubic Zr₅₀Ti₃₅Nb₁₅ medium-entropy alloy with unique properties has been developed in Ref. [29]. This Zr₅₀Ti₃₅Nb₁₅ shows a unique combination of Young’s modulus (62 GPa), yield strength (657 MPa), tensile ductility (22%), and excellent corrosion resistance.

11.2 Applications of additive manufacturing (AM) technologies

3D layer manufacturing technologies are expected as new technologies for manufacturing acetabular cups, artificial hip joint stems, and femoral components and tibial trays of artificial knee joints among others. 3D layer manufacturing technologies are also expected for manufacturing porous materials such as acetabular components. It is possible to obtain marketing approval for highly biocompatible implants that are optimized for the skeletal structures and needs of patients by combining 3D layer manufacturing technologies with imaging technologies such as CT. 3D layer manufacturing technologies using Co-28Cr-6Mo alloy powder can be also applied to the manufacture of humeral components of artificial shoulder joints, metacarpal and proximal phalanx components of artificial finger joints, tibial and talar components of artificial ankle joints, and femoral components of artificial knee joints.

Additive manufacturing (AM) by selective laser melting is used to fabricate 3D objects in a single stage directly from their computer-aided design (CAD). In a conventional casting process, materials constantly expand and contract during the formation of wax patterns, investment, and casting. Relatively dense femoral components free of blow holes are formed with AM technology, reducing the occurrence of problems such as the breakage of femoral components caused by casting defects. In selective laser melting, a laser is scanned along metal powders on the basis of slice data to obtain a layer of a product. The powder for the next layer is placed on the melted layer, and the laser is again scanned on the basis of the data for the next slice [15]. The development of orthopedic implants customized to the skeletal structure and symptoms of each patient is now possible.

12. Conclusion

Zirconium (Zr), niobium (Nb), and tantalum (Ta) are effective alloying elements for Ti alloys to achieve long-term superior biological, biochemical, and biomechanical compatibilities. The oxides ZrO₂, Nb₂O₅, and Ta₂O₅ strengthen the passive (oxide) films and prevent metal ion release in long-term clinical use. Ti-15Zr-4Nb alloy has excellent mechanical compatibility (room temperature strength and fatigue properties). Indeed, the biological compatibility of this alloy has been proven in cytotoxicity, sensitization, irritation, systemic toxicity, genotoxicity, and rat and rabbit implantation tests. This has led to high expectations for the long-term future use of this alloy as a viable implant material. The durability limit of various types of osteosynthesis devices linearly increases with increasing bending strength. The relationship (durability limit at 10⁶ cycles) =0.67× (bending strength) (N·m) (R² = 0.85) is obtained for bone plates, spinal rods, intramedullary nail rods, CHSs, short femoral nails, and epiphyseal plates by regression. This slope of 0.67 is close to the ratio of the fatigue strength to the tensile strength of the raw metals. The relationship for the highly biocompatible Ti-15Zr-4Nb alloy is also linear. The fatigue strength of the Ti-15Zr-4Nb alloy stem hot-forged at 780°C is 850 MPa, which is higher than that of the Alloclassic SL Stem. Selective laser melting is a promising new manufacturing technology for artificial hip stems and artificial knee femoral components, and artificial joints. A bearing with a cross-linked UHMWPE (XLPE) insert shows a low wear rate (5 mm³/Mc), suggesting its excellent long-term clinical performance.

The oxide coloring of a Ti materials surface is globally used in osteosynthesis devices. As shown in Figure 26, the correlation of color tone and oxide film thickness (Y, nm) can be determined with respect to anodic oxidation potential (X, V), yielding a linear correlation given by Y = 0.286 + 1.521 [X], R² = 1.0 [30].

Figure 26.

Correlation between oxide film thickness and anodic oxidation potential.

Owing to the rapid progress of 3D layer manufacturing technologies, the development of orthopedic implants customized to the skeletal structure and symptoms of each patient is now possible.

Conflict of interest

The authors declare no conflict of interest.