TITANIUM . 99 SCIENCE AND TECHNOLOGY

Corrosion Resistance, Mechanical Properties, Corrosion Fatigue Strength and Biocompatibility of New Ti Alloy without V for Medical Implants

Y. OKAZAKI*, Y. ITO** and E. NISHIMURA***

* Mechanical Engineering Laboratory, Agency of Industrial Science and Technology, Ministry of International Trade and Industry, 1-2 Namiki, Tsukuba, lbaraki 305-8564, Japan

**Titanium Metals Technology Department, Kobe Steel, Ltd, 1-Chome, Chiyoda-ku, Tokyo 100-8221, Japan

*** Japanese Industrial Standards Center, Agency of Industrial Science and Technology, Ministry of International Trade and Industry, 1-2 Namiki, Tsukuba, lbaraki 305-0044, Japan

Corrosion resistance due to wear changes with the materials used as disk and pin, frictional load, potential zone and with the pH of the solution. A new Ti-15%Zr-4%Nb-4%Ta alloy containing 0.2%Pd showed a excellent wear access corrosion properties compared to other Ti alloys. The addition of 0.2%0 and 0.05%N to the new Ti alloy and heat treatment increased the ultimate tensile strength to 1000 MP a The total elongation and reduction of area was more than 10% and 50%, respectively. The effect of the load profile estimated by human hip joint simulation on corrosion fatigue strength was compared with the sine wave. The fatigue strength of the new Ti alloy to which 0 and N were adled under a sine wave of 10 Hz at 108 cycles was 900 MPa The effect of the frequency on the fatigue strength at 2 Hz and 10 Hz was almost similar. Moreover, the fatigue strengths between the sine wave and human hip joint load profile were almost the same for the new Ti alloy. The relative growth ratios of fibroblasts L929 and osteoblastic MC3T3-El cells with new Ti alloy plates were slightly higher than those for Ti-6%Al-4%V ELI (Extra Low Interstitial) alloy. The relative growth ratios of L929 and MC3T3-El cells were close to l (non-cytotoxicity) in the wear test using the apatite ceramics pin on the new Ti alloy disk up to 105 wear cycles in Eagle's MEM solution. However, the relative growth ratio decreased sharply for the Ti-6%Al-4%V ELI alloy disk against the apatite ceramics pin from 3xl0

4 wear cycles since V ions were released

during wear into the wear test solution because the pH of the Eagle's MEM solution rose as the number of wear cycles increased. Further, the new Ti alloy implant has a new bone tissue formation rate that is similar to Ti-6%Al- 4%V ELI alloy when implanted in rat tibia up to 29 Ms (48 weeks).

Key words: Ti alloy, corrosion resistance, mechanical property, corrosion fatigue property, biocompatibility, wear access corrosin test, cell culture test, rat tibia implantation, released metallic ion.

1. INTRODUCTION

The population of persons over 65 years of age is increasing every year in Japan, and the use of artificial body implants is increasing accordingly. Along with various implant materially such as Co-Cr alloy and SUS316L stainless, the use of Ti-6%Al-4%V ELI alloy as a body implant alloy is increasing. /3 type Ti-15%Mo-5%Zr-3 %Al and a+ 13 type Ti-6%Al-2%Nb-l %Ta alloys have been developed for artificial hip joints by Kobe Steel Ltd in Japan and are clinically used as artificial hip joints for cement and cementless types, respectively1

'2

'3

• For cementless artificial hip joints made of Ti-6%Al-2%Nb-l %Ta alloy, a part of the neck in the stem surface was improved with titanium plasma spray coating in combination with a bottom coating of apatite-wollastonite containing glass-ceramic to obtain a stable born contact sooner after implantation. Hydroxyapatite-roated artificial hip joints made of Ti-6%Al-4%V alloy have also been developed using an inert-gas shielded arc spray technique with titanium powder in combination with hydroxyapatite-coating by Kyocera Corporation

4• The

effects of various metallic concentrations on the relative growth ratios for cultured cells were examined using extracted mediums with various metallic particles. Ti, Zr, Nb and Ta show low cytotoxicit/'

6• As a + /3 type

alloys for medical implants not containing V, Ti-5%Al-2.5%Fe and Ti-6%Al-7%Nb alloys are specified by ISO 5832-10 and 5832-11 standard<;. A near /3 type alloy, Ti-13%Nb-13%Zr alloy, is also specified in ASTM F 1713-96 standard /3 type alloys having a slightly lower Young's modulus than that of a+ /3 type alloy, such as Ti-12%Mo-6%Zr-2%Fe7

, Ti-15%Mo-3%Nb-3%Al 8, and Ti-29%Nb-13%Ta-5%Zr

9 alloys, are also being

developed for medical implants. Ti-13%Nb-13%Zr alloy is reported to possess a higher adhesion of osteoblasts anda lower bacterial adhesion than pure Ti and Ti-6%Al-4%V alloy10

: Our'group has reported on the effects of Zr, Nb, Ta and Pd on corrosion resistance in a physiological saline solution, mechanical properties and biocompatibility with a.iltured cells (cytocompatibility)11

'12

'13

• Corrosion resistance has been improved by adding Zr, Nb, Ta, and Pd because the resultant Zr02, Nb20 5, Ta20 5 and PdO strengthen the Ti02 passive film formed on new alloy 14

• The room-temperature mechanical strength of the annealed (973K - 7.2 ks) alloy was increased by adding Zr and small quantities of oxygen and nitrogen. As a consequence, Ti-15%Zr-4%Nb-4%Ta-0.2%Pd alloy

1135

TITANIUM. 99: SCIENCE AND TECHNOLOGY

showed excellent corrosion resistance, mechanical properties and biocompatibility compared to Ti-6%Al-4%V ELI alloy. Wear-accelerated corrosion of Ti alloys in the biological environment using cyclic polarization measurement from 0 to 5 V has been reported15

• a + {3 type Ti-6%Al-4%V and Ti-6%AI-7%Nb alloys possessed the best combination of rorrosion and wear resistance, although rommercially pure Ti and the near {3 type (Ti-13%Nb-13%Zr) and {3 type (fi-15%Mo) alloys displayed excellent rorrosion resistance. In this work, the corrosion resistance in a physiological saline solution, the mechanical properties and the biorompatibility using ailtured cells and rat tibia implantation for Ti alloys were examined The effect of heat treatment for new Ti alloy, namely, Ti-15%Zr-4%Nb-4%Ta containing 0.2%Pd, 0.2%0 and 0.05%N, on mechanical properties and corrosion fatigue properties were also investigated The fatigue test was carried out under a tension-to-tension mode with a sine wave at a stress ratio of 0.1 and at a frequencies of 2 Hz and IO Hz in Eagle's medium solution using a rorrosion-fatigue cell with 90%N2+5%C02+5%02 gas bubbling. Further, the effect of the wave shape on corrosion-fatigue strength was examined using a sine wave and load profile estimated by an analysis of forces and actions of the human hip joint.

2. MATERIALS AND EXPERIMENTAL METHOD

2.1. ALLOY SPECIMENS AND HEAT TREATMENTS

The new Ti~15%Zr-4%Nb-4%Ta containing 0.2%Pd, 0.2%0, 0.05%N (hereafter called Ti-15%Zr-4%Nb-4%Ta), pure Ti grade 2, Ti-6%Al-4%V ELI, Ti-6%Al-2%Nb-1 %Ta and {3 type Ti-15%Mo-5%Zr-3%Al alloys were melted by vaaium arc melting. In the case of a+ 13 type alloys, after 13 and a - {3 forging, the alloys were annealed for 7.2 ks at 973 K. For /3 type alloy, solution treatment was carried out at 1058 K for 1.8 ks and then quenched in water. Moreover, to determine the optimum solution treatment ·and-aging conditions· for the Ti-15% Zr-4%Nb-4%Ta alloy, alloy specimens 17 mm in diameter and IO mm in thickness were ait after a-13 forging. These specimens were kept in a range of 1028 to 1073 K (755 to 800 'C) for 3.6 ks and then water cooled The solution treated specimens were aged in a range of 623 to 723 K (350 to 450 'C) for 18 to 54 ks (5 to 15 h) and then air cooled Table 1 shows the chemical composition of these materials. For comparison, the chemical composition of Co-Cr alloy and SUS316L stainless steel is also shown in Table 1.

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

Cr Ni Mo Mn c Si Fe Co

SUS316L 16.31 12.21 2.05 0.22 0.013 0.23 Bal.

Co-Cr alloy 28.66 0.26 5.97 0.55 0.28 0.81 0.38 Bal.

Titanium alloys Zr Sn Nb Ta Pd Al v Mo Fe 0 N H c Ti

Pwe Ti grade 2 0.062 O.Q78 0.00370.0012 0.007 Bal.

Ti-6%Al-4%V EU 6.42 4.19 0.198 0.101 - 0.0052 0.011 Bal.

Ti-6%Al-2%Nb-l %Ta 1.91 0.95 5.90 o.n 0.214 0.055 0.0047 0.0010 <0.005 Bal.

Ti-15%Mo-5%Zr-3%Al 5.12 2.92 15.19 0.028 0.099 0.009 0.0073 0.007 Bal.

Ti-15%1.r-4%Nb-4%Ta 15.24 <0.01 3.90 3.92 0.22 0.022 0.162 0.04800.0110 0.002 Bal.

2.2. ANODIC POLARIZATION TESTS UNDER STATIC AND FRICTIONAL CONDffiONS

When the airrent density measured by anodic polarization testing is low, the corrosion resistance of an alloy is excellent. Several block specimens of 10 mm x 10 mm x 5 mm in thickness were ail from the sample alloys under static rondition. The surfaces of the specimens, except for 1 qn2

, were coated with epoxy resin and then polished with waterproof emery paper up to #600 under running water and ultrasonically cleaned in ethanol. The anodic polarization tests were ronducted independently in 1 mass% lactic acid (pH=2.5) and Eagle's MEM (Minimum Essential Medium: mainly composed of NaCl : 6.8 g, C6H120 6 : 1 g, KCI : 0.4 g, CaC12 : .0.2 g,

NaH2P04 : 0.115 g, containing various inorganic salts and kanamycin, etc. in 1 dm3 of pure water) containing 10% fetal bovine serum, 7.5%NaHCO (2.3 vol%), and 3%L-glutamine (1 vol%) (hereafter called Eagle's medium solution, pH=8.2"8.8) solutions. The testing solution was bubbled with high-purity nitrogen gas at a rate. of 3.33 x 10:6 m3/s (200 cm3/min) for 0.6 .ks (10 min). The specimen wa.<; initially held at -1 V for 300 s. An anodic polarization test was carried out from -1 V to a maximum of 5 V at a sweep rate of 6.67 x 10-4 V/s (40 mV/min) under a small amount of high-purity nitrogen gas flowing to the solution surface. The micromotion between the bone and the implant gives rise to metallic. ion release. This results in the loosening of the stem and thus giving rise to pain .. The effect of frictional load on the anodic polarizalion

1136

TITANIUM. 99: SCIENCE AND TECHNOLOGY

properties in 1 mass% lactic acid and Eagle's medium solutions wa<> investigated. A polarization cell to measure tht: anodic polarization curves under the frictional condition is shown in Fig. 1. The specimen electrore is reciprocally moved by a cam in the oontact with apatite ceramics. The applied load was displayed digitally by the load cell and the frictional load was adjusted with screws. To maintain the testing solution at 310 K, hot water wa<> cireulated around the cell. The reciprocal speed was varied by changing the gears and the ratio of teeth in the gear box. The cell is mare of teflon, and the holder to fix the ceramics is made of polychloro tetrafluro ethylene (PCTFE, Daiflon). A test specimen 10 mm in diameter and 10 mm in thickness was prepared out of sample alloy. After connecting the test specimen to a platinum wire by lug terminal and screw, they were inserted into daiflon, and the specimen suiface, except for 0.8 cm2

, was covered with epoxy resin. Thereafter, the specimen suiface was polished with waterproof emery paper up to #600 unrer running water, followed by ultrasonic cleaning in ethanol. The reciprocal distance was 5 mm, apatite12 and alumina ceramics were used for the friction specimen to simulate bone tissue, and the data collecting interval was 2 s, the reciprocal speed at 10·3 to 10·2 m/s (0.1 to I Hz), and the frictional load wa<> up to. 59 N. However, because the frictional area in this test was less than 10% of the surface area, the measuring area was a<>sumed to be a constant area of 0.8 cm2

•

After setting the specimen electrode and the friction

Mqtor

C.E.

C. E. : Counlar electrode (Pl) W. E. : WOfking aledrocla R E. : Ralarence electrode

Fig. I Schematic diagram of polarization cell for anodic polarization test under the frictional condition.

specimen in the polarization cell, the testing solution was d!aerated with high-purity nitrogen at a rate of 1.67 x 10-6 m3/s for 0.6 ks. The specimen wa<> initially held al -1.5 V for 300 s. An anodic polarization test wa<> then carried out from -1.5 V to 5 Vat a sweep rate of 6.67 x 10-4 V/s (40 mV/min) with ·a the small amount of high-purity nitrogen gas flowing to the solution suiface. Moreover, during the anodic polarization test, measurement was peiformed by adjusting the screws to minimize frictional load fluctuation (to a maximum 40% of the mean frictional load).

2.3. TENSILE TEST AND MICROSTRUCTURE OBSERVATION

Tensile testing was oonducted with test specimens 6 mm in diameter and 22 mm in gage length at room temperature at a aosshead speed of 8.33xlff6 mis (0.5 mm/min). After etching the test specimen in HF: HN03 :

H20 at a ratio of 15 : 20 : 65 cm3 containing 10%H20 2, the microstructure wa<> observed with an optical microscope, a scanning electron microscope (SEM) and a transmission electron microscope (TEM). The sample specimen for TEM observation was prepared by electrolytic polishing (35 to 55 V, 40 to 100 mA) in a mixture of 95% acetic anhydride and 5% perchloric acid solutions.

2.4. CORROSION FATIGUE TEST IN PYSHIOLOGICAL SOLUTION

The fatigue test specimen with the shape and dimensions shown in Fig. 2 was cut from the sample alloys. The corrosion fatigue test was carried out in Eagle's medium solution. To remove the inner strain

o_,

(40) 60 40

140

in the surface of the alloy caused in the manufacturing process, the surface of the test specimen was fully finished with #600 waterproof emery paper in the direction parn.llel to the test specimen. The test specimen was fitted up in a cell (polyethylene, inside

Fig. 2 Dimensions of test specimen for fatigue in physiological saline solution.

1137

TIT/\Nll IM· 99: SCIENCE /\ND TECHNOLOGY

diameter: 40mm, outside diameter: 50 mm, (a)

height: 55 mm) containing Eagle's medium, and the test specimen was then set on a corrosion fatigue testing machine. The corrosion fatigue test was amducted by circulating hot water around the cell containing the Eagle's medium to maintain the insici: temperature at 310 K. To make the oxygen concentration in the saline solution closer to that of a living body and to maintain pH at 7.4, a small amount of 90%N2+5%C02+5%02 mixed gas was bubbled into the medium solution. As fatigue test conditions, a sine wave having a stress ratio (R=(minimum tensile stress) I

0 0.2 0.4 0.6 0.8 1.0 Time,t/s

(maximum tensile stress)) of 0.1, a frequency of 10 and 2 Hz, and a number of cycles up to 108 times were set. Moreover, fatigue tests using a load profile estimated Fig. 3 The equipment used for fatigue test with load profile from the analysis of forces and actions for estimated by analysis of forces and actions for human hip joint. the human hip joint17 (called the hip joint (a) 858 Mini Bionix, (b) Physiological saline solution cell, load profile) were also conducted. ( c) Hip joint load profile. Figure 3 shows the MTS model 858 Mini Bionix equipment, cell and the hip joint load profile used in the test.

2.5. CYTOCOMPATIBILITY

Two types of cells, namely, L929 cells chived from murine fibroblastic tissue and murine osteoblast-like MC3T3-El cells were used The culture medium for the L929 cells was prepared by adding 7.5% NaHC03

solution (2.3 vol%), fetal bovine serum (10 vol%) and 3% L-Glutamine solution (1 vol%) to the Eagle's MEM solution. For the MC3T3-E1 cells, the culture medium was prepared by actling fetal bovine serum (10 vol%) and 7.5% NaHC03 solution (2.4 vol%) to a- MEM solution. Alloy plates (33 mm in diameter and 1 mm in height) were cut from the sample alloys. The surfaces of the plates were then polished with waterproof emery paper up to #1000 unrer running water and ultrasonically cleaned in ethanol. All the plates were sterilized in an autoclave at 394 K (121 'C) for 1.8 ks. The plates were then placed in a culture dish 35 mm in diameter. 2.8 ml medium solution and 0.2 ml cell suspension containing 3.0xl0

4 L929 and 5.0xl04 MC3T3-El cells were separately

seeded on these plates. These dishes were then incubated for 345.6 ks (4 d) under a 95%air-5%C02 atmosphere at 310 K (37 'C). After incubation, the cells were completely separated by pipetting using trypsin (0.1 mass%, 3 mL) and electrolyte (balanced electrolyte solution, 3 mL) and transferred into a Coulter cup containing 3 mL Eagle's medium for L929 cells or a - medium for MC3T3-El cells. The cells were then counted using a Coulter counter. The number of cells in each dish in 0.5 mL of medium containing trypsin and electrolyte was amnted four times and the average number of cells in each dish was estimated The Ti-6%Al-4%V ELI alloy plate was the control. The relative growth ratios of the L929 and MC3T3-El cells were estimated using the following formula: (average number of cells per dish after 4 d incubation ) I (average number of cells in the control). The average and standard deviation values of the relative growth ratio were estimated using more than 5 dishes. Metallic ions are released from Ti-6%Al-4%V alloy implants insire the human body 16

• To acrelerate the release of metallic ions, wear tests were carried out using Ti-15%Zr-4%Nb-4%Ta alloys, pure Ti grade 2 and Ti-6%Al-4%V ELI alloy disks (5 mm in thickness and 70 mm in diameter) against these alloys and apatite ceramics pin (20 mm in length and 9 mm in diameter) in Eagle's MEM solution (pH: 4.3) not containing fetal bovine serum

and 7.5% NaHC03 solution. The wear tests were carried out under a load of 9.8 N, a rotation diameter of 27 mm, a rotation speed of 0.2 mis up to 105

cycles. The experiment was carried in a cell made of polycorbonale material. After wear testing, Eagle's MEM solution wa<; centrifuged at 50 s·1 for 0.9 ks, and the supernatant wa<; filtered with 0.2, 0.1, 0.05, 0.025, and, finally, a 0.015 /1 m membmne using a vacuum pump to remove wear particles. Thereaf1er, Eagle's medium was prepared by adding fetal bovine serum (10 vol%), 7.5% NaHC03

solution (2.3 vol%) and 3% L-Glutamine solution (1 vol%) to this Eagle's MEM solution. 3.0xl04 L929 and 5.0xl04 MC3T3-El cells were seed:d using this worn Eagle's medium and a - medium amtaining wom Eagle's medium, respectively. The metallic concentration in this worn medium wa<; analyzed by !CP-MS (inductively coupled plasma mass spectrometer). Eagle's medium without wearing and a- medium not containing the worn Eagle's medium were used a<; control for L929 and MC3TI-El cells, respectively.

1138

TITANIUM . 99: SCIENCE AND TECHNOLOGY

2.6. CHEMICAL ANALYSIS OF METALLIC ELEMENTS IN MEDIUM BY ICP-MS

Perfluoro alkoxyl alkane (PFA) flasks were cleaned with ultra-pure water(18.3 M Q) after use and filled with a 5 mac;s% nitric acid solution and carefully plugged to exclude air. The flask was emptied before use, and cleaned inside and outside with a 5% nitric acid solution and ultra-pure water. A Yokogawa AnalyticaJ Systems HP4500 spectrometer with auto sampler (CETAC ASXSOO) was used in this study. The mass axis and intensity were aqusted with. a tuning solution containing 10 mass ppb Li, Y, Ce,· and Tl. The typicaJ measurement conditions for plasma and the autosampler are as follows: frequency: 27.1 MHz., radio­frequency (R.F.) power: 1.2 kw, plasma gas (Ar) flow rate: 15 ckn3/min, auxiliary gas (Ar) flow rate: 1.0 dm3/min, carrier gas (Ar) flow rate: 1.1 dm3/min, sampling distance: 5 mm, rinsing rate by peristaltic pump with 3% nitric acid and ultra-pure water: 0:5 s·1 (2x10·3 dm3/min), rinsing time: 180 s, exchanging rate by peristaltic pump for sampling solution: 0.5 s·1, exchange time for sampling solution: 120 s, stabiliz.ation time for peristaltic pump: 60 s, simple uptake rate: 4.Sxl0-4 dm3/min (peristaltic pump speed: 0.12 s"1

).

The isotopic mac;s numbers used for meac;urement of Ti , Al and V concentrations are Ti: 49 and 50, Al: 27 and V: 51, respectively. The integration time for the mass number is 1 to 3 s. The micro-pipette tips for sampling were cleaned with a 5% nitric acid solution and ultra-pure water before use. The single-element stan<hrd solutions (SPEX CertiPrep, Inc., 1000 mass ppm) were diluted to use as the Ti, Al and V stan<hrd solutions. Ultra-high-purity nitric acid (f AMAPURE -AA-10) wac; used for preparing meac;urement solutions. The 0.5 mass ppb Cobalt or Indium were used ac; the internal stan<hrd solution: The working curves were established in the concentration of 5 points and above. · All ICP-MS measurements were taken in a clean room (clac;s 10000) and the measurement solutions prepared in a clean bench (clac;s 100). The internal standard method is effective for correcting equipment fluctuation in long-term meac;urements. A standard solution was adcW to the Eagle's and a - medium solutions so that they contained 100 mac;s ppb of Ti, Al and V concentrations. The medium solution was diluted with 1 % nitric acid solution from 2 to 500 times to compare Ti, Al and V concentrations determined by ICP-MS. The result is shown in Fig. 4. Twice dilution may result in damage to the instrument since the medium solution contains a high level of salts. So, the Ti, Al, and V concentrations in the Eagle's and

120 ..--------------..

110

:c 100 a. a.

~ .5. 90 (.)

80

li

(a) Eagle's n-.m 100 maaa ppb Tl. Al, v in Eagle's medium

ln1amal standatd : Co

0 0: Tl

0 :Al l:l: v

0

~ 70"-~ ............... ~~~....._~~~~ ~ 120 1~ ___ 10 ____ 100 ___ __,1000

::li: c

~ i 110

~ g

100 ~ ::li:

90

80

70

(bl a -medium 100 mass ppb Ti. Al, v 1n a- medium

lnlemal standard : Co

10

0: Tl

0 :Al l:l: v

100

Oilutlon Factor

1000

Fig. 4 Effect of dilution for Eagle's medium (a) and a - medium (b) containing 100 mass ppb Ti, Al and V on the metallic concentration measured by ICP-MS using Co as internal standard. :c a. a. 1000 .---------------,,,.

cri. 100 ::li: ti. !:2

a- medium solutions are approximately measurable when the medium solution is diluted by 5 to 50 times. Ti, Al and V ~ 10

::> stan<hrd solutions were a~ to the Eagle's medium to obtain ~

various concentrations. These medium solutions were then ~

diluted to 25 times with 1 mass% nitric acid solutions. Figure · j 5 shows comparisons of Ti, Al and V concentrations measured 8 by ICP-MS. The determination for Ti, Al and V 8 concentrations at roughly more than 1 mass ppb was possible by directly diluting Eagle's and a - medium solutions from

10 100 1000

Metallic Concentration Added to Medium, C (mass ppb) 10 to 40 times with 1 mass% nitric acid solutions. However, these concentrations were changed by the existence of hydrochloric and sulfuric acid solutions, and use of the Pyrex glass flask.

Fig. 5 Comparison of the metallic concentrations measured with 49Ti, 27Al and 51 V using Co as internal standard.

1139

TITANIUM· 99: SCIENCE AND TECHNOLOGY

2.7. RAT TIBIA IMPLANTATION

Six-week-old maJe Wistar rats which were fed for more than 1 week and weighed 104-124 g were used. Implant specimens 2.0 mm in diameter and 1.5 mm in height for Ti-15%Zr-4%Nb-4%Ta and Ti-6%Al-4%V ELI aJloys were rut from the sample aJloy by considering the siz.e of tibia. The edges of the column were then rut (C 0.1) and the surfare was polished with waterproof emery paper. After pentobarbitol sodium solution (nenbutaJ solution) was intraabcbminally injected in a cbse of 25 mg/kg, the site IO mm below the knee joint

on right and left sides wac; shaved, sterilized and incised with a scalpel. The bone surfare was exposed and isolated to the bone membrane. For the implant cavity, a hole 2 mm in diameter and 2 mm deep wac; fonned perpendicular to the longer axis of the cortical bone with an infusion of sterile physiological saJine. For this, round <kntal burs 1 mm (#3) and 2 mm (#8) in diameter were used in tum using a miaomoter for implant surgery rotating at low speed. The Ti-15%Zr-4%Nb-4%Ta and Ti-6%AI- 4%V ELI aJloys were then implanted into the bone marrow of the left and right sides of the tibia, respectively, and the wound was sutured Tibiae; on the right and left sides were extracted from 5 rats for the preparation of <kcalcified tissue (n=3) and un<kcalcified tissue (n=2) samples after death by an intraab<hminal administration of pentobarbital sodium solution 6, 12, 24 and 48 weeks after implantation. The proredure for the preparation of <kcalcified

(a) tissue sample is shown in Fig. 6 (a). Bone cuuingdirection

tissue was fixed with a 10% neutral fonnalin solution and decalcified with a 10% formic acid formalin solution. After <kcalcifying, bone tissue was rut with a blade parallel to the longer axis of tibia and so as if to divide the vertical axis of the Ti alloy implant into two. Then, the Ti alloy implant wac; removed The bone tissue that was ail into two was dehydrated and defatted with 70-100% alcohol and 100% aretone. After embedling with paraffin wax, embedded specimen were slired to a thickness of 4 µ. m using a miao rutting machine. These slired sections were stained with Hematoxylkin Eosin stain (H. E. stain) and Az.an Malloy stain (A M. stain). The structural changes- of new bone tissues were observed with an optical microscope. The proredure for the preparation of undecalcified tissue sample is also shown in Fig. 6 (a). Bone tissues were fixed as above. Tissues embedded in ~in were ail to a thickness of 100 µ. m parallel to the long axis of the tibia so as to divide the vertical axis of the Ti alloy implant into two with a diamond blade. The sliced resin block was then polished to a thickness of 40 µ. m with waterproof emery paper from #600 to #1000, followed by staining with H. E. stain. After H. E. staining, the structural changes of new bone tissues were observed. Furthermore, for the measurement method of thickness for new bone tissue as shown in Fig. 6 (b), the thickness of new bone tissue at IO sites formed in medullary was measured with <kcalcified and un<kcalcified optical micrographs for the central part of the Ti alloy implant. Changes in thickness in new bone tissues were statistically analyz.ed with Student's t-test.

(b)

Upper Rat tibia

Fixaiin l~lution

Decalcified in 10% formic acid Dehydration in 70-100% ethanol

001"~~--@--t > ""'"" =-11~ ~;Jj Bone Surface . Bone .

~ e: Re~oval and dehydration · ~ ~ Slicing of resin bl~k

~ • > """"'" • I 111oo • .Lfu-"'1 I ne t I ~

Slicing of paraffin wax Polishing of resin block slice

Bo~ p,raffin wax . Bone Ti Resin }I, " .:;

Staining

Decalcified section

I I I I I

1

2

3

Ti Implant

~ Staining

Undecalcified section

Cortical bone

Newly formed bone 10

9

8

I Mesial wall~$$$?/..$~ I I

I Cortical bone 4 5 6 7 Bone marrow

Fig. 6 Schematic diagram of the preparation of the microscopic sections (a) and measuring points for bone contact thickness (b).

1140

TITANIUM. 99: SCIENCE AND TECHNOLOGY

3. RESULTS AND DISCUSSION

3.1. CORROSION RESISTANCE

Figure 7 shows anodic polarization curves measured in 1 ma.5s% lactic acid and Eagle's medium solutions at 310 K. A passivation peak for pure Ti grare 2 and Ti alloys is not observed and the polarization curve immediately enters the passivity zone. The current rensity in the potential region of I V and over increac;es as the potential increases, and increases rapicDy in the potential region of 2 V and over, exrept for /3 type Ti-15% Mo-5%Zr-3% Al alloy. Figure 8 shows the change in the width of the frictional area after anodic polari?.ation test in the Eagle's medium solution. The widths of frictional area for the Ti alloys are larger than those of SUS316L stainless steel and the Co-Cr alloy. However, the effect is small when the mean frictional load is 20 N or more. Figure 9 shows oomparisons of anodic polarization curves unrer a static oondition and the mean frictional load of 49 N at a frequency of 1 Hz with apatite ceramics in the Eagle's medium solution at 310 K (37 'C ). The current density was higher than that of static state and fluctuates by destruction and formation of passive film. In the ea.5es of SUS316L stainless steel and the Co-Cr alloy, fluctuation of the current density is seen only in the passivity zone, but almost no fluctuation is seen in the active and

Sr--~-~---r--~-------,7_,----,

(a) I% Lactic acid I ; / ~·-.'·' pH=2 .++-o :o·~'+;.-- _ ~-

~ ~~,.. ~- .. -"'-4 ... "I>· ~~,.. :./ L Pure Ti grade 2

.;,¥' -<.• I _.!;.-:-.- . . .. .~ · _.,..---.::-·· Ti-6%A!-4%V ELI •.. ----·· 3 ( )--~ r __ ... --·········-·'-

,..-:' \ .---

w u UJ

2

3

2

0

r" I \ .. ---------)__,..; ____;., - T~15%Mo-5%Zt-3%AI

, .. LCO-Cr

<:~::.·:.f"SUS316l ·········-·-.

-1 [_ _ _.___.__,__._ .......... ~ __ ..___.__._ ................... ~---'-~~5 O.Q1 0.1 1

Current density , I I A • m·2

Fig. 7 Comparison of the anodic polarization curves for various implant alloys in deaerated 1 mass% lactic acid (a) and Eagle's medium (b) solutions under the static condition at 310 K.

transpassive zones. As to the Ti alloy, fluctuations of the current density arising from friction are seen from the activity zone to the high potential region, increasing in fluctuation widths of current density (maximum- minimum current densities) and the mean current rensity (the mean value of maximum and minimum values). In the case of Ti-15%Zr-4%Nb- 4%Ta alloy a.5 shown in Fig. 9 (d), the fluctuation of the current rensity is small up to the high-potential zone. To examine the effect of pH, the anodic polarization test was oonducted in a 1 % lactic acid solution with a mean frictional load of 20 N. Figure 10 shows the result of the Ti-6%Al-4%V ELI and Ti-15%Zr-4%Nb-4%Ta alloys. There is a tendency for the current density to increase in comparison with the results of the Eagle's medium solution. Figure 11 shows the effect of reciprocal speed on anodic polarization curves of Ti-15%Zr-4%Nb-4%Ta alloy with a mean frictional load of 20 N. The effect of friction is

E WI-" 0 2 4 6 0 5,......;:..-~--r--~~.--~--r-, -"' Apatite i:eramics

8: Ti-15Zl-4Nt>-4Ta

0 20 40 60

Mean Frictional Load , WI N

small irrespective of increases in the reciprocal speed Figure Fig. 8 Change in the width of frictional area 12 shows changes in the corrosion potential we to friction of after anodic polarization with apatite ceramics 1 Hz. The SUS316L stainless steel and the Co-Cr alloy show in deaerated Eagle's medium solution at 310 K a small decrease in corrosion potential in spite of the increased as a function of mean frictional load. frictional load On the other hand, the Ti alloys show considerable decreases.

I 14 I

TITANIUM· 99: SCIENCE AND TECHNOLOGY

s..--~~~~~~~~~~~~~~~~-,

~4 en

- 1 (ti

'E 0 .S! ~-1

w () 4 en

(a) SUS316L stainless steel

O.D1

I I

/

Q1 1 ro Current density , I IA • m·2

100

5

4

3

2

0

·1

5 . {d) Ti-15%Zr-4%Nb-4%Ta

4

3

2

+-49N 0

·1

·2 0.01 0.1 1 10 100

Current density , I IA • m·2

Fig. 9 Effect of kinetic frictional force of 49 N on the anodic polarization curves with apatite ceramics for various implant alloys in Eagle's medium solution at 310 K. (a) SUS316L stainless steel, (b) Co-Cr alloy, (c) Ti-6%Al-4%V ELI, (d) Ti-15%Zr-4%Nb-4%Ta alloy.

(a) Ti-6%Al-4%V ELI

w 0 w 0

~ ~ ~ ·1 ~ -1 1;,...-..

~ .2......,~_._ .................. ~_._........_......,...__._........_......,...,___._ ................ w > UI 100 Uj ·2 0.01

Oi "" ~ 4 11.

2

0

·1

-2~~_._........_ ....... ~_.._......._.....,......__._........_......,...__._ ................. w

0.1

(a) 1o"m·s·1

Apali1e ceramlcs

Eagle' s medium 20N

10 100

0.1 1 10 100 0.01 0.1 1 10 100 Current density , //A· m·2

Current density , //A• m·2

Fig. IO Effect of kinetic frictional force of 20 N on the anodic polarization curves with apatiteceramics for Ti-6%Al-4%V ELI (a) and Ti-15%Zr-4%Nb-4%Ta (b) alloys in 1% lactiC acid solution at 310 K.

Fig. 11 Effect of lateral reciprocal speed on the anodic polarization curves for Ti-15%Zr-4%Nb-4%Ta alloy in Eagle's medium solution at 310 K.

1142

TITANIUM · 99: SCIENCE AND TECHNOLOGY

From this result, it is consid:!red that under the frictional environment, the stressing zone turns anodic and its periphery cathodic, and corrosion tend> to progress compared to the static environment. In order to compare the effect of friction among the alloys, the effect of friction was examined by dividing the mean value of the current d:!nsity by a static value of current d:!nsity at eadt potential. The results are shown in Fig. 13. The increa<>e in the current density due to friction with the Co-Cr alloy is small ~

100 ..---------------..

(a) Eagle's medium Apalite ceran>cs

0

compared to the Ti alloys. With respect to the Ti alloys, there is ~

a tend:!ncy for the effect of friction to increase in the low 8 potential region; in the high potential region, however, the effect ~ of friction is small. Aluminum ceramic pin instead of apatite s ceramic pin also show the same tendency.

w ·0.4 u en .; > > -0.6

~ -c 0

ti -0.8 'C ... "' c ·c

-1.0 ::> 'ti iii = c ~ -1.2 ~ c .2 ., g ·1.4

8

0

0

2

20

W/kg 4 6

Apatite ceramics

10·2m-s "1(1Hz)

I : s.o. of Ee

40 60

Mean Frk:tional Load . WI N

c ~ 0.1 -c ·1 0 2 ... 100 ..---------------.. 0 .g IU a: ~ .. c ~ c ~ ::> u

(b) 1% Lactic acid

10

Pureng.-2

0.1 .__~_._~~~~---'-~""--J ·1 0 3

Potential , EI V vs. SCE

4

Fig. 12 Change in the corrosion potential during friction with apatite ceramics in deaerated Eagle's medium at 310 Kasa function of mean frictional load.

Fig. 13 Change in the ratio of current density for friction to static conditions as a function of potential for apatite ceramics in Eagle's medium (a) and 1 mass% lactic acid (b) solutions.

3.2. MECHANICAL PROPERTIES

The Ti-15%Zr-4%Nb-4%Ta alloy specimens were kept in the 1028 to 1073 K (755 to 800 'C) for 3.6 ks, water cooled, then aged at 673K (400 'C) for 36 ks. Figure 14 shows changes in the 0.2% proof strength (a o.2%PS), ultimate tensile strength (a UTS), total elongation (f.E.) and rediction of area (R. A) of the new Ti alloy as a function of solution treatment temperature. The 0.2% proof strength, ultimate tensile strength, total elongation and rediction of area OOc::rease with a higher solution treatment temperature. The reduction of area, in particular, d:!creases sharply with a lower vol% of a phase as shown in Fig. 14. The Ti-15%Zr-4%Nb-4%Ta alloy was solution-treated at 1048 K (775 'C) for 3.6 ks and aged Figures 15 (a) and (b) show changes in the 0.2% proof strength, ultimate tensile strength, total elongation and rediction of area as a function of aging temperature and aging time; respectively. Strength and d.Jctility are affected very little by aging temperature and aging time. We assumed the optimum heat treatment conditions for the Ti-15%Zr-4% Nb-4%Ta to be aging at 673K for 29 ks (8 h) after solution treatment at 1048 K for 3.6 ks.

8!. 1.2 ~--------~ Cl -5 0

1.0

f ~ 0.8

0

0.6

~ ~·

1020 1040 1060 1080

Solution Treating Temperture, TI K

100 -l. .. ..

80 IU s:: a. 0

60 0 .,. 40 ~

l ui

20 i-:

l 0 <

Fig. 14 Effect of solution-treating temperature for Ti-15%Zr-4%Nb-4%Ta alloy on the mechanical properties at room temperature and volume percentage of a phase.

Figure 16 shows a comparison of the TEM microstructures for Ti alloys. Ti alloys annealed at 973K for 7.2 ks are mostly lath of a' martensite, while the solution-treated Ti-15%Zr-4%Nb-4%Ta alloy consists of the primary a phase and a' martensite. Fine a phase precipitate is due to aging after solution treating. A good

1143

TITANIUM. 99: SCIENCE AND TECHNOLOGY

balance between strength and ductility is derived when the fine a pha<ie is unifonnly distributed by solution treatment and aging. Comparisons of the mechanical properties of Ti alloys at room tempernture are summarized in Table 2 .

., a. 1.2 CJ

(a) -!/) ... :>

OUTS 0 ., 1.0 -------a. 00.2%PS CJ o---0-° !/)

100

80

0.. ,,.

~ "' 0 0.8 0

60

40

0.6 T. E. Q---o---0

0 ~ <i.

0.4 a: 550 600 650 700 750 800

Aging Temperture, TI K .,

1.2 a. CJ (b) -u; ... :> 100

0 ., 1.0 a. 80 CJ !/) 0..

•• ~ 0.8 0

A. A.

ts. ~ 60

40

~ 0.6 T. E.

u.i 20 ~

0-----0---o

0.4 L.-~-'--~-.L..-~-..___. 0 20 40 60

Aging Time, t I ks

Fig. 15 Effect of agmg temperature (a) and aging time (b) forTi-15%Zr -4%Nb-4%Ta alloy after solution treatment at 1048 K for 3.6 ks on the mechanical properties at room temperature.

(f) Ti-15%Zr-4%Nb-4%Ta 973- 7.2 ks Annealing

(h) Ti-15%Zr-4%Nb-4%Ta S. T. + 673 K- 29 ks Aging

tt2 • :rz,,30 0: ol1 • •

• aoo -• 011 on

12' 'I e e •

1• )1(511)

(e) 111

011. 100 • 000 • foo e Oti • 111.

• a.(011)

(g) Ti-15%Zr-4%Nb-4%Ta 1048 K -3.6 ks Solution treating

; , -~-:, Fig. 16 TEM micrographs for ., :~.. :.,,.-co·· various Ti alloys after annealing

" " • .,..~ ·:C.~-- at 973 K for 7.2 ks and solution • "' ,, · ;' treating plus aging. (a), (b) and

t ,...,-,,.~-.. "' (t) annealing, (c) solution treating -~~ ,;.' at 1058 K for 1.8 ks, diffractions

t ·, in bright field ( d) and dark field ~ ~ ~ « (e) in (c), (g) and (h) S.T. plus aging.

Table 2 Comparison of the mechanical properties for various Ti alloys at room temperature.

a 02.,,PS/ MPa Ours/MPa T.E.(%) A. A.(%) EIGPa

Ti-6Al-4V ELI Annealing 869.:!.4 921!6 22! 1 52! 6 101! 5

Ti-6Al-2Nb-1 Ta Annealing 936 !l 964.:!.0 17 ! l 45! () 125 ! I

Ti-15Mo-5Zr-3Al S.T. 838.:!.2 852!2 25!1 48! 1 80!0

Ti-15Zr-4Nb-4Ta Annealing 877 !7 881!4 27!4 59!1 100± I

S.T.+Aging 918 !l 1026! 3 14!0 51!1 97 !l

Ti-6AI· 7No* Annealing 800 900 IO 25

•Minimum value in ISO Standard

1144

TITANIUM. 99: SCIENCE AND TECHNOLOGY

3.3. CORROSION FATIGUE STRENGTH

Figure 17 (a) shows a romparison of the S-N cmves obtained from the rorrosion fatigue test with the sine wave in the Eagle's medium solution. The fatigue strength at the 10

7th cycle for 8 type alloy,

which have romparatively high strength at room temperature, was lower than that of a + 8 type alloy annealed at 973 K. The number of cycles to failure for Ti-15%Zr-4%Nb-4%Ta alloy annealed at 973 K inaea-;ed as maximum stress was decreased, and it was found that rupture stress at 108 cycles c:oincick!s at approximately 600 MPa The fatigue strength of Ti-6%Al-4%V ELI alloy at the 107th cycle is also romparatively high. The fatigue strength ofTi-6%Al-2%Nb-1%Ta aJloy at the 108th cycle was approximately 700 MPa In the fatigue­fractured surface of a + 8 type alloys, fatigue cracks originated from stress roncentration regions such as inclusions in the periphery of the fatigue test specimen. A number of fatigue cracks starting from the inclusion and a cleavage type fracture mod! with river pattern were seen on the fracture surface of the 8 alloy. Comparison of S-N curves obtained from the rorrosion fatigue test using a sine wave and human hip joint load profile after solution treatment

.. 0... :::i:

~ b .,; .. ., t. UJ E ::J E ~ :::i:

1200

1000

800

600

'400

200103

1200

1000

800

600

(a) Annealing Sine wave R=0.1

0 10 Hz and 2 Hz

D 0 a+ f3 type

Q:T•llAHVEU o: Tl-6Al-2 H1>-1Ta ().: Tl-15Zr -4NJ>.4Ta fl.: Ti-15 Mo-5Zr ~ : Co-Ct alloy

10• 105 10 8 107 10

(b) Solutlon treallng+Aging for TI-15Zr-4Nb-4Ta

• 0 :Sine wave

10 Hz. R=0.1

e : Hip joint profile

2Hz

Cycles to Faliure, Nf (Arbitrary unit)

109

plus aging is shown in Fig. 17 (b). The fatigue strength is higher than with annealed specimens. Also, the effect of fre<p.iency on fatigue strength al 2 Fig~ 17 Comparison of S-N curves obtained by fatigue test Hz and 10 Hz were almost similar, and the fatigue with sine wave for various Ti alloys annealed (a}, and effect strengths were the same for the sine wave and of wave shape with hip joint load profile and sine wave human hip joint load profile for Ti-15%Zr-4%Nb- on S-N curves for new Ti alloy aged after solution treatment 4%Ta alloy. (b) in a physiological saline solution at 310 K.

3.4. CYTOCOMPATIBILITY FOR Tl ALLOYS

Figure 18 shows a romparison of Ti aJloy plates for the relative !!J.

growth ratios of L929 and MC3T3-El cells. The relative growth ~ 1.2

ratios of L929 and MC3T3-El cells for the Ti-15%Zr- w

4%Nb-4%Ta alloy are slightly higher than those ofTi-6%Al-4%V ~ u 1.0

alloy (rontrol). The change in the relative growth ratio of L929 ~ cells as a function of wear cycles is shown in Fig. 19 (a) for the ! o.s

wear test with alloy disk against the apatite ceramia; pin from 104 ~ to Ht wear cycles. The relative growth ratio remained at 1 ~ 0_6

0: L.929CeR

• :MC3T3·E1

~

(non-cytotoxicity) for pure Ti grad! 2 and Ti-15%Zr-4%Nb-4%Ta ~ alloy. Figure 19 (b) shows the change in the relative growth ratio 1 o.• of MC3T3-El cells with a - medium oontaining worn Eagle' o medium prepared for 4xl0

4 and 10

5 cycles of wear in Eagle' MEM ~ 0.2

solution. The relative growth ratio of MC3T3-El cells sharply £ ~ed in the case of Ti-6%Al- 4%V ELI alloy at both 4xl04 and

O Tl-6Al-4V Pure Ti THW-2Nb Tl-15Mo- ll·15Zr-4Nb 105 wear cycles, while the relative growth ratio remained at 1 for pure Ti grad! 2 and Ti-15%Zr-4%Nb-4%Ta alloy. Figures 20 (a) and (b) show the changes in the ooncentrations of Ti, Al and Vin Eagle's medium anda - medium solutions as a function of wear cycles and a volume perrentage of worn Eagle's medium acli:d to a- medium, respectively.

EU ~2 ·1Ta 5Zr·3AI -4Ta

Fig. 18 Comparison of cytocompatibility on the various Ti alloy plates with fibroblast L929 and osteoblastic MC3T3-El cells.

Figure 20 (a) shows that the concentrations of Ti and Al in the worn Eagle's medium solution were almost the same as the number of wear cycles increased, but the metallic concentration of V ions in the the wear tested

1145

TITANIUM. 99: SCIENCE AND TECHNOLOGY .

Eagle's medium solution gradually increased as the number of wear cycles increased from 104 to Hf cycles. The metallic concentration of Al in a - medium gradially increased, but the V concentration increased sharply. The effect of V concentration on cell viability obtained by the extraction of metallic particles in Eagle's and a - medium5

'6 is

summarized in Fig.21. A good coincidence between Fig. 20 and Fig. 21 for the V ion concentration was observed The various metallic concentrations for the Ti-15%Zr-4% Nb-4%Ta alloy in Eagle's medium after 105 wear cycles were very low (fi < 0.01, Zr : 0.004, Nb : 0.001, Ta : 0.0004 and Pd : 0.01 mass ppm). The pH of the Eagle's MEM solution increases as the wear cycles increac;e clue to wear of the alloy disk against the apatite ceramics pin. Based on these results, it is thought that as the number of wear cycle increases, pH increases and so the release of V ions increases in the case of wear of the alloy disk against the apatite ceramics pin. However, as the' effect of AI concentration on cell viability depends on surface roughness, surface treatment, strength of the AI oxide film and extracting condition with Al particles19

, we suggest that further research is needed on the cytotoxicity of Al.

0.6 (a) Eagle's medium

0.5

e o.4 c. c.

0

.!!! a; (..) CD

"' ~ 0 0

~ a:

~ (!)

"' ~ .. a:

.!!! a; (.)

w I'.! "' (..) ::!: 0 .Q 10 a: J::.

~ (!) .. 2: N .. a:

1.2 TI-15Zt-4Nl>-<4Ta

t.O

0.8

0.6

0.4

TI-6Al-4V EU 0.2 0

0

0 2 4 8 ·to t2

Number of Wear Cycles, NI Ht

t.2 Ti·15Zt-4Nt>-4Ta

1.0 1-1'.'.l~.-ft./\r----W-----f) ·10• Cycles

0.8

0.6

0.4

0.2 4K104 Cycles (9

0 0 20 60 80 100

Percenlage of Eagle's medium in a medium (vol'lb)

a 0.3 ..s u

Fig. 19 Change in the relative growth ratio of L929 cells with the number of wear cycles in Eagle's medium

,; 0.2 E :I

1i 0.1 ::!:

solution (a) and the volume percentage of worn Eagle's medium for alloy disk against apatite ceramics in a -medium.

,, 0

iii !" .. !

0 2 6 8 10 12

Number of Wear Cycles, N f 1 r1 .s 0.2 .----------------, c

i g 8 .Y

~ ::!: 0.1

0

(b) Cl medium

1o•Cydes in Eagle's

MEM-n

0 20 40 60 80 too

Percenlage of Eagle's Medium in a medium (vol'lb)

Fig: 20 Change in the metallic concentration in Eagle's medium as a function of number of wear cycles (a) and with the volume percentage of worn Eagle's medium forTi-6%Al-4%V ELI disk against apatite ceramics pin in Eagle's MEM solution at 10 5

wear cycles in a-medium (b).

1.2

(a)

'·'I~••• .!!! 0.8 ~ Cytotoxic , .

~ L929 MC3T3·E1 L929 MC3T3·E1 Ti:Q () Nb: A A Zr: 0 WI Ta: e Q

w 0.6

~ ..., 0.01 O.t 10 ::!: t.2.----------------.

~ (b) CD

"' ~ 0

~ 0.8

~ 0.6 (!) .. ~ 0.4 a; a:

0.2

a.at 0.1

A: L929Cell

Q :MC3T3-E1

10

Metallic Concentration in Medium, C (mass ppm)

Fig. 21 Effects of Ti,Zr,Nb,Ta concentration (a) and V concentration (b) on the relative growth ratios of

. L929 and MC3T3-El cells.

1146

TITANIUM. 99: SCIENCE AND TECHNOLOGY

3.5. RAT TIBIA IMPLANTATION FOR TI ALLOYS

Representative undecalcified tissue specimens for Ti-15%Zr-4%Nb-4%Ta and Ti-6%Al-4%V ELI alloys after 7.3 Ms (12 weeks) implantation are shown in Fig. 22. Inflammatory cells and foreign body tissue response were not obseived. Decalcified tissue specimens stained with H.E. are shown in Figs. 23 and 24. From the these results, it is clear that new bone is formed around the alloy implant in the case of both Ti alloys. It can also be seen that no strange born formation was obseived.

,. (b)

Fig. 22 Optical micrographs of undecalcified sections of new Ti (a) and Ti-6%Al-4%V ELI (b) alloys stained with H. E. stain after 7.3 Ms (12 weeks) implantation .

(a) ].6 Ms (6 weeks) · (h) 7.] Ms ( 12 week:.)

Ti-IJ'iAl-·Vi V ELI

1100 um I

Fig. 23 Optical micrographs of dccalcified sections of Ti-6%Al-4%V ELI alloy stained with H. E. stain after 3.6-29 Ms (6-48 weeks) implantation.

1147

(a) 3.6 Ms (6 weeks)·

(c) I-Li M.,. (24 weeks)

' "\. "fi, • ~

TITANIUM. 99: SCIENCE AND TECHNOLOGY

... marrow

(b) 7.3 Ms (12 wecksf·------

' t l Ti- I VrZr-.+,..i-Nh--:VrTa

····-········· .. Bone nrn m.>w '(cl) 29 M~ (48 weeks)

Fig. 24 Optical micrographs of decalcified sections of Ti-15%Zr-4%Nb-4%Ta alloy stained with H .. E. stain after 3 .. 6-29 Ms (6-48 weeks) implantation ..

In addition, representative decalcified tissue samples at 29 Ms (48 weeks) are shown in Fig. 25. Since the A.M .. stain is stained blue, new bone tissue with mature calcification formed at 29 Ms after implantation ..

(.t)

. Bone marrow . .;. . )lj_""---41

... · ..S' ~Jli'· -~ r100 fem

• ··! l , oo i· m I

Fig. 25 Optical micrographs of dccalcified sections of new Ti (a) and Ti-6%Al-4%V ELI (b) alloys stained with A. M. stain after 29 Ms (48 weeks) implantation.

1148

TITANIUM. 99: SCIENCE AND TECHNOLOGY

E A romparison of the thickness of new bone tissues is shown in Fig. 26. In both alloys, the thickness of new bone tissue tended to increase from 3.6 Ms (6 weeks) to

.: 150 .------------------. Cl ,,;

14.5 Ms (24 weeks) and was almost unchanged after 14.5 c:

Ms. The results of the statistical significant difference test ~ Ql 100

showed no significant difference in the thickness of new E bone tissues with p<0.05 among 3.6, 7.3, 14.5 and 29 Ms ~

>. ( 6, 12, 24, 48 weeks) for both alloys. 1

4. CONCLUSION z 50 0 ::l Ql c: ""

0: Ti-15Zr-4Nb-4Ta 0 : Ti-6Al-4V

(1) The current density in the high potential region is very ~ 0 ~ ~~~~-__._ _ __._ _ _.__...._____.

low for the new Ti-15%Zr-4%Nb4%Ta alloy, thus which fii o 10 20 30

shows ronsiderably high rorrosion resistance. The rurrent ~ Implantation Time, t I Ms

density becomes higher during friction than during static Fig. 26 Change in the mean thickness of newly ronditions. The fluctuation widlh was observed in the fonned bone as a function of implantation time. passivity zone for Co-Cr alloy. However, in the case of Ti alloys, the fluctuation widlh was observed in both activity and passivity zones. Among the Ti alloys, wring friction against apatite ceramics, the ament density was low for the new Ti alloy in the high potential region, and the fluctuation widlh of the rurrent density was narrow rompared to other Ti alloys. The effect of the lateral speed was also negligible for the new Ti alloy rompared to other Ti alloys. For rorrosion resistance wring wear ronditions, dtange of the materials used as disk and pin, frictional load, potential mne and the pH of the solution, the new Ti alloy showed excellent wear rorrosion properties rompared to other Ti alloys. (2) As the solution treatment temperature increased,. the ductility of the new Ti alloy deaeac;ed. After solution treatment, the medtanical strength did not dtange mudt as the aging time and the aging temperature were increac;ed. The adlition of 0 and N to the new Ti alloys and heat treatment substantially increased the ultimate tensile strength to 1000 MPa, and total elongation was more than 10 %. The fatigue strength of /3 type Ti-15%Mo-5%Zr-3%AI alloy at 10

7 cycles was lower than that of a + 13 type alloys. The fatigue strengths of

Ti-15%Zr-4%Ta-4%Nb and Ti-6%Al-2%Nb-1%Ta alloys annealed at 973 K for 7.2 ks at 108 cycles were about 600 and 700 MPa, respectively. The.fatigue strength of solution-treated and aged new Ti alloy at 108 cycles was 950 MPa The effect of at 2, Hz and 10 Hz frequencies on the fatigue strength was almost the same. Further,

the fatigue strength was almost the same for the sine wave and human hip joint load profile. (3) The relative growth ratio of L929 and MC3T3-El cells for the new Ti alloys was slightly higher than that of Ti-6%Al-4%V ELI alloy plate. In the case of wear test in Eagle's MEM solution, the relative growth ratio remained at 1 for pure Ti grade 2 and Ti-15%Zr-4%Nb- 4%Ta alloy. On the rontrary, the relative growth ratio of L929 cells steeply decrea<ied from 3xl0

4 number of wear cycles and almost became equal to 0 at 105 wear cycles

for the Ti-6%Al-4%V ELI alloy disk against the apatite ceramics pin. The pH of the Eagle's MEM solution increac;es as. the wear cycles inaea5e for the wear of the alloy disk against the apatite ceramics pin. As t~.e

number of wear cycles increases, pH increases and so the release of V ions increases. (4) The new Ti alloy implant has a similar new bone tissue fonnation rate to Ti-6%Al-4%V ELI alloy when implanted in rat tibia The mean thickness of the newly formed bone around the new Ti and Ti-6%Al-4%V ELI alloy implants rontinuously increased up to 14.5 Ms (24 weeks), and dtange in the mean thickness thereafter, until 29 Ms (48 weeks} were very small.

REFERENCES

1. Y. MATSUDA, T. NAKAMURA, K. IDO, M. OKA, H. OKUMURA and T. MATUSHITA: "Femoral romponent made of Ti-15Mo-5Zr-3AI alloy in total hip arthroplasty", J. OTthopaedic Science, 1997, 2, p.166-170.

2. K. IDO, Y. MATSUDA, T. YAMAMURO, H. OKUMURA, M. OKAandH. TAKAGl: "Cementless total hip replacement", Acta Orthopaedica Scandinavica, 1993, 64, p. 607-612.

3. T. YAMAMURO, T. NAKAMURA, H. lIDA and Y. MATSUDA: "A new model of bone<.anserving cmentless hip prosthesis made of high-tech materials: Kobelco H-5", Joint ATthroplasty, ed. S. lMURA, M. WADA and H. OMORl, Springer-Verlag Tokyo, 1999, p. 213-224.

4. K. HAYASHI, T. lNADOME, T. MASHlMA, Y. SUGlOKA: " Comparison of bone-implant interface shear strength of solid hydroxyapatite and hydroxyapatite-roated titanium implants", J. Biomedi.cal Materials Research, 1993, 27; p. 557-563.

1149

TITANIUM. 99: SCIENCE AND TECHNOLOGY

5. Y. OKAZAKI, S. RAO, S. ASAO, T. TATEISHI, S. KATSUDA and Y. FURUKI: "Effect of metallic Ti, Al and Von cell viability", Materials Transactions, JIM, 1999, 39, p. 1053-1062.

6. Y. OKAZAKI, S. RAO, S. ASAO and T. TATEISHI: "Effect of metallic concentrations other than Ti, Al and Von cell viability", Materials Transactions, JIM, 1999, 39, p. 1070-1079.

7. K. K. WANG, L. J. GUSTAV ASON and J. H. DUMBLETON: "Miaostructure and properties of a new beta titanium alloy, Ti-12Mo-6Zr-2Fe, d:veloped for surgical implants", Medical Applications of Titanium and Its Alloys: The Material and Biological Issues, ASTM STP 1272, ed. S. A BROWN and J. E. LEMONS, American Society for Testing and Materials, 1996, p. 76-87.

8. S. K. BHAMBRI, R. H. SHElTY and L. N. GILBERTSON: "Optimization of properties of Ti-15Mo-2.8Nb -0.2Si & Ti-15Mo-2.8Nb:.0.2Si-260 Beta Titanium Alloys for application in prosthetic implants", Medical Applications of Titanium and Its Alloys: The Material and Biological Issues, ASTM STP 1272, ed. S. A BROWN and J. E. LEMONS, American Society for Testing and Materials, 1996, p. 88-95.

9. M. NIINOMI: "Mechanical properties of biomedical titanium alloys", Materials Scienre and Engineering A243, 1998, p. 231-236.

10. P. KOVACS and J. A DAVIDSON: "Chemical and electrochemical aspects of the biocompatibility of titanium and its alloys", Medical Applications of Titanium and Its Alloys: The Material and Biological

Issues, ASTM STP 1272, ed. S. A BROWN and J. E. LEMONS, American Society for Testing and Materials, 1996, p. 163-175.

11. Y. ·OKAZAKI, Y. ITO, A ITO and T. TATEISHI: "Effect of alloying elements on mechanical properties of titanium alloys for medical implants", Materials Transactions, JIM, 1993, 34; p. 1217-1222.

12. Y. OKAZAKI, Y. ITO, A ITO and T. TATEISHI: "New titanium alloys to be consid:red for medical implants" , Medical Applications of Titanium and Its Alloys:. The Material and Biological Issues, ASTM STP 1272, ed. S. A BROWN and J. E. LEMONS, American Society for Testing and Materials, 1996, p. 45-59.

13. 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. Biomedical Materials Research; 1993, 34, p. 1217-1222.

14. Y. OKAZAKI, T. TATEISHI and Y. ITO: "Corrosion resistanre of implant alloys in pseucb physiological solution and role of alloying elements in passive films", Materials Transactions, JIM, 1997, 38; p.78-84.

15. M. A KHAN, R. L. WILLIAMS and D. F. WILLIAMS: "In-vitro corrosion and wear of titanium alloys in the biological Environment", Biomaterials, 1996, 17, p. 2117-2126.

16. K. KONDO, M. OKUYAMA, H. OGAWA and Y. SHIBATA: "Preparation of high-strength apatite ceramics", Communications of the American Ceramic Society, 1984, 67, p. C-222-223.

17. L. C. MEJIA and T. J. BRIERLEY: "A hip wear simulator for the evaluation of biomaterials in hip arthroplasty components", Bio-Medical Materials and Engineering, 4, 1994, p. 259-271.

18. H. J. AGINS, N. W. ALOCK, M. BANSALl, E. A SALVATI, P. D. WILSON, P. M. PELLICCI and P. G. BULLOUGH : "Metallic wear in failed titanium alloy total hip replacement", Journal of Bone and

Joint Surgery, 70-A, 1988, p, 347-356. 19. Y. OKAZAKI, S. KATSUDA, Y. FURUKI and T. TATEISHI: "Effect of Aluminum oxid: film on

Fibroblast L929 and V79 cell viability", Materials Transactions, JIM, 1999, 39, p. 1063-1069.

·'

1150