Low wear rate of UHMWPE against zirconia ceramic (Y-PSZ) in comparison to alumina ceramic and SUS 316L alloy Praveen Kumar, Masanori Oka,* Ken Ikeuchi,+ Koichiro Shimizu, Takao Yamamuro; Hideo Okumura: and Yoshihiko Kotoura Department of Artificial Organs, Research Center for Medical Polymers and Biomaterials, ‘Department of Mechanical Engineering, and $Department of Orthopaedic Surgery, Kyoto University, 53, Kawahara-cho, Shogoin, Sakyo-ku, Kyoto 606, Japan Partially stabilized zirconia ceramic is being recognized among ceramics for its high strength and toughness. With this ceramic, is possible to manufacture a 22mm-size femoral head for low friction arthroplasty of the hip joint in association with an ultra-high-molecular-weight polyethylene socket. Wear-resistant properties of zirconia ceramic were screened on two principally different wear devices. Sterile calf bovine serum, physiological saline, and distilled water were chosen as the lubricant fluid media. Depending on the lubricant medium, the wear factor of polyethylene against zirconia ceramic
counterfaces was 40 to 60% less than that against alumina ceramic counterfaces, and 5 to 10 times lower than with the SUS316L metal counterfaces. Polyethylene wear against metal was more susceptible in saline in which it had 2 to 3 times higher wear rate than with serum. On the other hand, different fluid media had little effect on polyethylene wear against ceramic counterfaces. In each set of tests, the wear factor obtained on an unidirectional wear device showed 10 to 15 times higher values, in comparison to the wear factor estimated on a reciprocating wear device.
INTRODUCTION
All prostheses having bearing surfaces release wear products into joint cavities. The gradual accumulation of these debris particles generates granulomatous reactions in the tissues surrounding the prostheses, and may cause loosening and other problems.1’2High-density polyethylene bearing against metal surfaces normally release variable amounts of plastic and metal due to wear and corrosion. To cope with the problem of wear, Boutin has introduced an alumina/alumina ceramic combination for the socket and ball head in a total hip replacement. Several studies have shown that the wear rate of polyethylene against alumina ceramic is 2 to 3 times less than the wear rate of polyethylene against stainless-steel a l l ~ y . ~ Semlitch -~ et a1.I’ have determined that the wear rate of polyethylene against alumina ceramic is about 20 times lower compared to polyethylene against Co-Cr-Mo alloy. They be‘To whom correspondence should be addressed. Journal of Biomedical Materials Research, Vol. 25, 813-828 (1991) CCC 0021-9304/91/070813-16$4.00 0 1991 John Wiley & Sons, Inc.
KUMAR ET AL.
814
lieve this favorable tribological behavior of ceramic in contact with polyethylene may be due to better corrosion resistance, wettability with liquids, and scratch resistance of the ceramic materials compared to those of metallic implant materials. However, alumina ceramic exhibits a brittle tendency and is sensitive to microstructural flaws." Currently, partially stabilized zirconia ceramic is being recognized for its high strength (and surface finish, making this material potentially suitable for the highly loaded environments found in joint replacement.I2-l4Wagner,I5and o t h e r ~ ' ~have , ' ~ examined the biological reaction of zirconia ceramic in vivo, and have found no difference between its biocompatibility and that of alumina ceramics. However, very little is known about this material's wear resistance properties against UHMWPE. Therefore, the purpose of the present study is to evaluate the wear resistance properties of zirconia ceramic against UHMWPE in association with presently used alumina ceramic/UHMWPE and SUS316L/ UHMWPE combinations. Two principally different wear screening test devices, i.e., unidirectional and reciprocating, are used for this purpose. By using the two different devices, we will try to determine the probable mechanical influence on mechanism of wear, if any.
MATERIALS
Ultra-high-molecular-weight polyethylene (UHMWPE) UHMWPE polymer pins were prepared from Hoechst RCH 1000 surgical grade polyethylene manufactured by the Zeigler low-pressure polymerization process and supplied by Kyocera Corporation, Japan. The molecular weight quoted by manufacturers was in the range of 3.5-4.0 million and a specific gravity of 0.94-0.95. Ceramics and metals counterfaces Mechanical properties of high-density alumina ceramic and yttrium oxidepartially stabilized zirconia (Y-PSZ) ceramic as well as 316L stainless steel (SUS316L) metal alloy are noted in Table I. Y-PSZ ceramic has been manufactured by Kyocera Corporation, Japan, by sintering zirconia powder with 2.53.0 mol.% of yttrium oxide (Y,O,) at a temperature of 1300"-1400"C. Geometrical forms of polyethylene wear pins as well as counterfaces were different in the two wear devices (Fig. 1).Discs used for the unidirectional wear device were annular, having an outer diameter of 60 mm and inner polished diameter of 40 mm with a thickness of 10 mm. For the reciprocating wear device, a rectangular flat surface of 28 mm x 28 mm with a thickness of 10 mm was prepared. It was possible to polish the zirconia ceramic surface of the prosthetic component up to a surface roughness of 0.006 pmRa. For the comparative study, the counterfaces for both devices were polished up to the same surface roughness of about 0.006 pmRa in all three materials, i.e., zirconia ceramic, alumina ceramic, and SUS316L metal.
WEAR OF UHMWPE AGAINST ZIRCONIA
815
TABLE I Selected Mechanical Properties of Zirconia Ceramic, Alumina Ceramic, and SUS 316L Alloy
Properties
Bulk density (lo3 x kg/m2) Flexural strength (MPa) Ultimate tensile strength (MPa) Young’s modulus (GPa) Vickers hardness (MPa) Elongation at fracture (%) Average grain size (pm)
Zirconia Ceramic (Y-PSZ)
Alumina Ceramic
6.0 1000
3.9 500
-
380 17,500 2.5
210 12,500 0.3
SUS 316L Alloy
550 200 1,600 40
Distilled water, physiological saline, and filtered sterile calf bovine serum were chosen as lubricant fluid media. The bovine serum for all sets of tests was from a single lot supplied by CC Laboratories, Cleveland, Ohio, lot number 00846-01 and stored at -20°C until use. Before each test, 30 mg of cefazolin sodium was added in 100 mL of bovine serum to prevent serum degradation by bacteria. MET HODS
Test conditions in the two devices were different, as listed in Table 11. All the polyethylene wear pins used in the devices were 5 mm in nominal thickness and machined from a single block of UHMWPE. For each set of tests, a fresh polyethylene specimen and counterface were used. The tests were repeated three times for zirconia and alumina ceramics and two times for Load
Load
Figure 1. Specimen geometry used in two wear devices: (A) reciprocating wear device, (B) unidirectional wear device.
KUlvlAR ET AL.
816
TABLE I1 Test Conditions in Two Wear Devices Test Conditions
Reciprocating Wear Device
Unidirectional Wear Device
63.58 3.45
12.56 3.00
50
60
Contact Surface Area (mm’) Contact Stress (MPa) Sliding Speed (mm/s.) Temperature (“C) Room Lubricant medium
25
- 26
25
37
- 26 -.
SUS316L metal. Cleaning of the polyethylene pins and counterfaces was done according to the ASTM F732 protocol. Unidirectional wear device (Figure 2) An unsterilized polyethylene cylinder, 4 mm diameter, with a contact surface area of 12.56 mm2 was pressed endwise under a constant axial stress of 3 MPa against the flat surface of the counterface disc. The disc was mounted in a plexiglass chamber having the capacity to contain 40-50 mL of lubricant fluid medium. The wear chamber was driven in a clockwise direction by a variable speed electric motor at a constant speed of 60 mm per second. Wear rate was evaluated by the decrease in polyethylene pin length, and frictional LOADING ARM
U N T E R WEIGHT
LE(ICL4SS CHAMBER MEDIUM
I
> 1
I
c
. I
Figure 2. Outline diagram of unidirectional wear and friction testing machine. Wear rate was estimated by a decrease in polyethylene thickness and frictional force by deformation of the leaf spring through capacitance type displacement probes.
WEAR OF UHMWPE AGAINST ZIRCONIA
817
force by deformation of the leaf spring. Both values were estimated with the help of capacitance type displacement probes (ST-0503, Iwatsu Electric Co., Japan), with gap adjustor (GA-202, Iwatsu Electric Co., Japan). These probes were capable of measuring the displacement with an accuracy of less than 2 pm. These probes were mounted over the machine and connected with transducers (ST-3501, Iwatsu Electric Co., Japan) and a continuous recording chart. Creep deformation and wear volume were recorded without removing the load at any stage. Room temperature of between 24" and 26°C was kept constant throughout the test. Before running the wear chamber, the polyethylene specimen was kept under an initial constant axial stress for a period of 12-24 h or more. During that period, in the first few hours, the polymer deformed rapidly at a rate of about 5 to 10 pm per hour and then at a much slower rate. In the last 3-4 h of loading, when the recording chart showed a steady state, i.e., further creep did not appear on the chart paper, the wear chamber was driven continuously for at least 1 week or more depending on wear prominence.
Reciprocating wear device (Figure 3) Preparation of polyethylene pins and other test conditions used here followed the ASTM F732 protoc01.'~An unsterilized polyethylene cylinder of 9 mm in diameter with a contact surface area of 63.56 mm2was mounted on the wear machine and pressed endwise against the flat counterface surface in a plexiglass wear chamber under constant axial stress of 3.45 MPa. The wear chamber was driven through a 25-mm stroke, or at the sliding distance of 50 mm per cycle with a frequency of 60 cycle per minute. The drive speed, and room temperature of 24"-26"C were kept constant throughout the test. A thermocouple and a heater, connected with a programmable control device,
LOADING ARM
Figure 3. Outline diagram of reciprocating wear and friction testing device. Wear rate was estimated using the same displacement probe used in the unidirectional wear device. Frictional force was measured by strain gauge sensor attached to the leaf spring.
KUMAR ET AL.
818
were inserted into the wear chamber to maintain the temperature of the lubricant f h i d medium at 37°C. Frictional force between the polyethylene specimen and the counterface was monitored by a strain gauge fixed on a leaf spring attached to the transverse bar holding the polyethylene pin. Wear rate was measured by the decrease in wear pin length using the same capacitance type displacement probe used in the unidirectional wear device. To exclude the creep deformation from apparent wear, the polyethylene specimen was treated similar to the unidirectiona1 wear device before running the wear chamber. To prevent dust or other contaminants, both wear devices were fully covered with a transparent polypropylene throughout the test. Compensation for fluid evaporation was made by siphoning distilled water into both devices. Wear factor and coefficient of friction were calculated as follows (Dowson et al."): Wear factor = Wear volume(mm3)/Load(N) x Sliding distance(m) Coefficient of friction = Friction force/Total load Surface changes on polymer pins and counterfaces were analyzed before and after each test under a stereoscopic optical microscope and scanning electron microscope (SEM), as well as visually. Surface topography of both the polyethylene wear pins and the counterfaces was also evaluated on a profilometer before and after each set of tests.
RESULTS
Each set of tests on the reciprocating wear device was allowed to run a total sliding distance of about 65 km or 1.3 million cycles (one cycle = 50 mm) and on the unidirectional wear device for about 30 -40 km continuously. Wear factors for all sets of tests are summarized in Table 111. UHMWPE wear against the SUS316L counterface was 2 to 15 times higher when compared against ceramic counterfaces. Between the two ceramics, the wear factor against zirconia ceramic was about 50% less than the alumina ceramic. The lubricant fluid media had little effect on wear of polyethylene against ceramic counterfaces, but had greater effect against SUS316L metal. With serum lubrication, polyethylene wear against the SUS316L counterface was 2 to 3 times less than in saline. The wear factor obtained in the unidirectional wear device showed a higher trend when compared with the wear factor obtained in the reciprocating wear device. The wear progress of polyethylene as a function of cycles in each set of tests is illustrated in Figure 4. The coefficient of friction between polyethylene and the counterfaces in each set of experiments are summarized in Table IV. An initial high coefficient of friction (0.01 to 0.02 above the mean value during the steady state) against the two ceramic counterfaces was recorded in the first hour of sliding in all three media used. This decreased gradually, stabilized after 2-3 h, and remained close to the values given in Table IV throughout the remainder of
10.7 t 12%
18.2 ? 6% 27.7 ? 30%
3 2
Unidirectional 7.5 t 3% 32.7 2 7 % 90.5 t 40%
1.01 ? 8% 1.81 t 4%
Unidirectional
0.57 3.89
? k
2% 8%
0.45 t 5%
Reciprocating
Saline
0.56 t 14%
Reciprocating
Bovine Serum
3
*Average and range.
Zirconia ceramic Alumina ceramic SUS 316L
Counterfaces
Specimen Number
‘Wear Factor (mm3/N . m) x
TABLE 111 Wear of UHMWPE on Two Different Wear Devices
11.8 ? 4% 37.1 t 10%
8.61 t 11%
Unidirectional
0.68 t 4% 1.12 +. 10%
0.38 -+ 6%
Reciprocating
Distilled Water
F
s o^z
bJ
5e
2
r,
9
m
I CJ
s3
KUMAR ET AL.
820
1.5
I.:\
o.20
vs. alumina c e r a m i c
E 0.25
2. a 2
0.5
0.75
1.0
X106
A
1.25
r 1.5 X106
vs. SUS 316L
2.4-
E
E
2.0
5 0
1.6-
’
1.2,
0.8 0.4 r
0.25
0.5
0.75
1.0
1.25
1.5
X106
Cycles
(C) Figure 4. Wear of UHMWPE as a function of cycles (one cycle = 50 mm), using bovine serum lubrication and a reciprocating wear device. Counterfaces: (A) zirconia ceramic, (B) alumina ceramic, (C) SUS316L.
the test. With the SUS316L counterface, the coefficient of friction in serum lubrication showed a similar trend as with ceramics. However, in saline and distilled water lubricants, the coefficient of friction against SUS316L metal increased (up to 0.1 to 0.3 above the mean value during the steady state) as the test progressed. In the unidirectional wear device, this was accompanied by an increase in lubricant temperature of about 2”-5”C above room temperature. After remaining at a peak for 3-4 h, the coefficient of friction fell gradually to the values shown in Table IV and remained constant for the remainder of the test, except for temporary rises several times during the test. Lubricant temperature also fell and remained steady at about O.Y-1”C above room temperature for the remainder of the test.
0.049 2 8%
0.056 f 10% 0.078 f 22%
3
3 2
Unidirectional 0.082 f 10% 0.115 f 20% 0.156 ? 17%
0.054 f 11% 0.065 f 18%
Unidirectional
0.075 f 2% 0.097 f 8%
0.089 f 10% 0.123 f 5%
5%
0.055
?
Unidrectional
0.044 2 5% 0.061 2 8%
0.028 2 8%
Reciprocating
Distilled Water
0.060 2 3%
Reciprocating
Saline
0.040 f 15%
Reciprocating
Bovine Serum
*Average and range of mean value during steady state.
Zirconia ceramic Alumina ceramic SUS 316L
Counterfaces
Specimen Number
*Coefficient of Friction
TABLE IV Coefficient of Friction of UHMWPE on Two Different Wear Devices
KUMAR ET AL.
822
On visual and stereoscope microscopic examination of surfaces of the two ceramic counterfaces, after the completion of wear tests in three lubricant media, not much difference in surface smoothness compared to their initial surfaces was demonstrated. On the saline- and distilled-water-lubricated SUS316L counterfaces, a pattern of surface scratches running the length of the contact area was seen as the test progressed. By the end of the test there was some evidence of polymer transfer under a stereoscopic microscope (Fig. 5). However, the transfer layer usually was not observed on the serumlubricated SUS316L counterfaces as the metal surface was clean after the test. Profilometer tracings of both ceramic counterfaces as well as SUS316L counterfaces, before and after tests, are shown in Figure 6. In the cases of ce-
Figure 5. Stereoscopic optical microscopic appearance of SUS316L counterfaces against UHMWPE after 1.3 million cycles on a reciprocating wear device: (A) in saline, (B) in distilled water. Counterface sliding direction was horizontal. Before test
in dist.water
In saline
In serum
Zirconia ceramic
Figure 6. Profilometer roughness tracings of the zirconia ceramic, alumina ceramic and SUS316L counterfaces against polyethylene. Tests were performed on a reciprocating wear device. (A, B,C) before test, (U, E, F) in distilled water, (G, H, I) in saline, and (J, K, L) in serum.
WEAR OF UHMWPE AGAINST ZIRCONIA
823
ramic counterfaces, little or no change from their initial surface roughness was recorded after the tests. On the other hand, SUS316L counterfaces became roughened five- to sevenfold from their initial surface roughness in saline and in distilled water, but not in serum lubricant (Fig. 6 F, I, L). The appearance of the worn polyethylene surface, examined with a scanning electron microscope, varied for each of the counterfaces (Fig. 7). Initially the surfaces were covered with ridges formed in the machining process (Fig. 7A). Against zirconia ceramic counterfaces, these machine marks became flattened but still visible over the entire surface indicating that very little wear had occurred (Fig. 7B). Against alumina ceramic counterfaces, the machine tracks were almost invisible in most of the areas and polymer debris adhered to the surface in some places (Fig. 7C). In contrast, after wearing out all machine marks on SUS316L counterfaces a new wear track was established on which islands of redeposited polyethylene were observed (Fig. 7D). The test conditions and duration of test were similar in each set.
DISCUSSION
Wear characteristics of two ceramics and SUS316L metal in association with UHMWPE have been studied. Results presented in this paper showed the advantage of ceramics over a steel counterface, where the wear rate of UHMWPE is decreased by about 5 to 10 times. For the two ceramics, polyethylene wear against the zirconia ceramic is 40 to 50% less than that against the alumina ceramic. Wear estimation
Laboratory screening tests are useful in determining the wear resistant properties of any new prosthetic material prior to clinical use. However, the procedure of measuring accurate wear rate is a controversial subject because of the effect of creep deformation on polyethylene height, and fluid absorption effects on polyethylene weight. McKellop et al.19 and Weightman et aL7 have successfully measured the wear rate by weight loss method. We also attempted to follow the weighing method described in the ASTM F732 protocol with a Sartorius 1-883 weighing machine by weighing the presoak control specimens and the specimens after the completion of tests. The polyethylene specimens were presoaked in serum or water (whichever was used in the wear test) for 10 to 15 days. The steady-state rate of weight gain of presoaked polyethylene specimens was about 2 to 4 pg per day. However, we could not get the accurate wear data (Table V). It is difficult to determine the exact cause of net weight gain presumably due to the greater fluid absorption during the wear test. It may be because of (a) effect of temperature on fluid absorption as the presoak specimen was stored at 4°C to prevent the biodegradation of lubricant medium and the experiment was performed at 37°C; (b) less wear of polyethylene against ceramic counterface which will
824
KUMAR ET AL.
Figure 7. Scanning electron microscopic appearance of UHMWPE surface: (A) initial wear face showed ridges formed in the machining process, (8)on completion of test against zirconia ceramic counterface (original machining marks still visible), (C) on completion of test against alumina ceramic counterface (original machining marks have been almost worn out and surface looks smoother), (D) on completion of test against SUS316L counterface showing original machining marks worn out and formation of a new wear track on polyethylene surface. Note: (B,C, D) after completion of 1.3 million cycles on reciprocating wear device in bovine serum lubrication.
need a prolong sliding distance to give a measurable weight loss; (c) some unknown biochemical and biomechanical mechanism between the polyethylene, counterface and lubricant medium during the wear test which may increase the fluid absorption.
WEAR OF UHMWPE AGAINST ZIRCONIA
825
TABLE V Comparison of Weight Loss of UHMWPE Estimated by Height Loss and Actual Weight Loss
Counterfaces Zirconia ceramic Alumina ceramic SUS 316L alloy
Height Loss Due to Wear (pm/1.3 x lo6 cycles)
Weight Loss Expected Due to Wear (mg/1.3 x lo6 cycles)
Actual Weight Loss Due to Wear (mg/1.3 x lo6 cycles)
14 13 11 20 23 24 42 39
0.836 0.777 0.657 1.195 1.375 1.434 2.510 2.331
-0.059, 0.298 -0.137) 0.388 -0.119, 0.165 0.370 0.836
*( -) Indicate weight gain of polyethylene specimen after completion of tests. Tests were performed on reciprocating wear device, with bovine serum as lubricant.
We found it more accurate and convenient to measure the thickness of polyethylene using a very sensitive displacement probe and gap adjustor capable of measuring with an accuracy of 2 pm. Before sliding, the polyethylene pin was pressed under constant load for a period till the recording chart did not show any further deformation for 3 to 4 h continuously. Thereafter, sliding was started from that point up to the end of the test. And because the decrease in polyethylene pin thickness was recorded continuously without removing the polyethylene specimen or stress during the test, it excluded the possibility of a decrease or increase in the polyethylene pin thickness due to creep deformation during the test, and recovery after the test from the actual wear. However, it was not possible to measure the creep deformation during the sliding which, if any, would continue to decrease.
Effect of lubricant media
The ASTM F732 protocol established bovine serum as an ideal lubricant medium for wear screening tests because of its similarity to normal joint synovial fluid. However, three lubricants were used in this study to determine their effect on polyethylene wear against three counterfaces. Results indicated that saline and distilled water showed different lubrication properties with the SUS316L counterface. In saline, the polyethylene wear against SUS316L metal was two or three times higher compared to bovine serum (Table 111).A transferred polyethylene layer on the SUS316L counterface was only visible when saline or distilled water was used as a lubricant. In bovine serum it was not visible. This may be due to serum proteins acting like a boundary lubricant, as suggested by McKellop” and others. Polyethylene transfer on ceramic counterfaces was not noticed in either fluid medium. Predictably, this may be due to the corrosion resistant properties of ceramic
KUMAR ET AL.
826
counterfaces and a better lubrication mechanism between the ceramic and polyethylene interface.
Two wear devices
In the present study, the wear factor recorded in the unidirectional device was 10 to 15 times higher than the reciprocating wear device results (Table 111). Though the reciprocating device is more relevant to prosthetic joints, the unidirectional device, because of its simplicity, is equally useful for comparative studies of various new and old prosthetic materials in different combinations over a short time span. Further, the severity of wear in the unidirectional device may be due to (a) reduction in the adhesive component of wear with reciprocating motion as suggested by Brown et a1.,2l (b) the ratio of area traversed by the polymer pin on the counterface to the area of the polymer pin, which is about 44 in unidirectional device, and about 4.5 in the reciprocating device. This may influence the wear mechanism in the two devices.
Surface finish It is generally believed, that the wear of UHMWPE decreased against both metal and ceramic counterfaces as the surface finish improved. Weightman7 reported that at a surface finish of about 0.025pmRa, the wear rate of polyethylene against alumina ceramic appeared to be the same as against stainless steel. According to his findings, the talysurf tracing of a metal surface contains more peaks and valleys which might cause an abrasive type of wear. On the contrary, the alumina ceramic surface has flat plateaus and this might be the reason for an adhesive type of wear. In the present study, the profilometric surface analysis did not show much difference among the three counterfaces, i.e., metal and two ceramics. However, SEM examination of zirconia and alumina ceramic surfaces showed great differences in their surface porosities (Fig. 8). This difference in porosity might be one of the reasons for the low wear of polyethylene against zirconia ceramic because, with the test progression, surface defects on the alumina ceramic would have to be filled slowly with transferred polymer.
CONCLUSIONS
(1) This study showed the wear resistant properties of zirconia ceramic against UHMWPE and confirmed the superiority of this ceramic over alumina ceramic and SUS316L metal alloy in terms of low wear and low friction. (2) Different lubricant fluid media had little effect on UHMWPE wear against ceramic counter faces but were prominent against the SUS316L metal. (3) Y-PSZ ceramic may be a biomaterial potentially suitable for low friction arthroplasty because of its better wear resistant properties and high strength.
WEAR OF UHMWPE AGAINST ZIRCONIA
827
Figure 8. SEM micrographs of polished surfaces showing porosity: (A) zirconia ceramic, (B) alumina ceramic. The authors would like to thank Mr. Makinouchi and Mr. Yanagida of Kyocera Co. Ltd., for their kind help in providing the materials and related information.
References 1. U. E. Pazzaglia, L. Ceciliani, M. J. Wilkinson, and C. Dell’Orbo, ”In2. 3.
4. 5.
6.
7. 8.
9.
volvement of metal particles in loosening of metal-plastic total hip prostheses,” Arch. Orthop. Truurnut. Surg., 104, 164-174 (1985). A. Pizzoferrato, L. Savarino, S. Stea, and C. Tarabusi, “Results of histological grading on 100 cases of hip prosthesis failure,” Biomuteriuls, 9, 314-318 (1988). P. Boutin, ”L‘alumina et son utilisation en chirugie de la hanche,” Presse Med., 79, 639 (1971). H. Mittlemeir, ”Selbsthaftende keramik-metal-verbund endoprostheses,” Med. Orthop. Technik, 95, 152-159 (1975). P. Griss and G. Heimke, ”Five years experience with ceramic metal composite hip endoprostheses 1 clinical evaluation,” Arch. Orthop. Truumut. Surg., 98, 157-164 (1981). H. McKellop, I. Clarke, K. Markolf, and H. Amstutz, ”Friction and wear properties of polymer, metal and ceramic prosthetic materials evaluated on a multichannel screening device,” J. Biomed. Muter. Res., 15, 619-653 (1981). B. Weightman and D. Light, “The effect of the surface finish of alumina and stainless steel on the wear rate of UHMW polyethylene,” Biomuterzals, 7, 20-29 (1986). H. Okumura, T. Yamamuro, P. Kumar, T. Nakamura, and M. Oka, “Socket wear in total hip prosthesis with alumina ceramic head,” in Biocerumic, Vol. 1, Proceedings Zst lnternutionul Bioceramic Symposium, H. Oonishi, H. Aoki, and K. Sawai (eds.), 1984, pp. 284-289. K. Kawauchi, Y. Kuroki, S. Saito, H. Ohgiya, S. Sato, S. Kondo, I. Hirose, s. Obara, H. Aso, K. Yamano, s. Sasada, and s. Norose, “Total hip endoprostheses with ceramics head and HDPE socket, clinical wear
KUMAR ET AL.
828
10.
11. 12.
13.
14. 15. 16.
17.
18. 19.
20.
21.
22. 23. 24.
rate," in Orthopaedic Ceramic Implants, Proc. Jap. SOC.Orthop. Ceramic Implants, H. Oonishi and Y. Ooi (eds.), 1984, 4, pp. 253-257. M. Semlitsch, M. Lehmann, and H. Weber, "New prospects for a prolonged functional life-span of artificial hip joints by using the material combination polyethylene/aluminium oxide ceramidmetal," J. Biomed. Muter. Res., 11, 537-552 (1977). J. P. Torre, "Manufacture and engineering properties of ceramics," Proc. Materials Eng. Conf., 5-7th November 1985, London, UK, Mechanical Engineering Publication Ltd., Bury St., Edmund, Suffolk, 1986. K. Shimizu, P. Kumar, M. Oka, Y. Kotoura, Y. Nakayama, T. Yamamuro, T. Yanagida, and K. Makinouchi, "Time dependent change of the mechanical properties of zirconia in vivo," Transaction of the 3rd. World Biomaterial Congress, 21-25th. April, 1988, Kyoto, Japan, Vol. 11, p. 406. K. Makinouchi, T. Yanagida, K. Shimizu, P. Kumar, Y. Kotoura, and M. Oka, "Development of zirconia ceramic for implants," in Orthopaedic Ceramic implants, Proc. Jap. SOC.Orthop. Ceramic Implants, H. Oonishi and Y. Ooi (eds.), 1984, 4. P. Christel, A. Meunier, M. Heller, J. P. Torre, and C. N. IPeille, "Mechanical properties and short term in vivo evaluation of Y-PSZ," J. Biomed. Muter. Res., 23, 45-61 (1989). W. Wagner, "Histologic bone reactions after implantation of alumina and zirconia pins," Trans. 7th European Conf. on Biomaterials, Amsterdam, 1987, p. 44. P. Kumar, K. Shimizu, M. Oka, Y. Kotoura, Y. Nakayama, T. Yamamuro, T. Yanagida, and K. Makinouchi, "Biological reaction of zirconia ceramic," Bioceruinic, Proceedings of 1st. International Bioceramic Symposium, Kyoto, Japan, 1988, Vol. 1, pp. 341-346. ASTM F732-82, Standard practice for reciprocating pin on flat evaluation of friction and wear properties of polymeric materials for use in total joint prosthesis, Annual book of ASTM Standards, ASTM, Philadelphia, 1982. D. Dowson, J. R. Atkinson, and K. J. Brown, "Advances in polymer friction and wear," Polymer Science and Technology Symposia Series, Vol. 58, L.-H. Lee (ed.), Plenum Press, New York, 1974, p. 533. H. McKellop, I.C. Clarke, K.L. Markolf, and H.C. Amstutz, "Wear characteristics of UHMW polyethylene: A method for accurately measuring extremely low wear rates," J. Biomed. Muter. Res., 1!2, 895-927 (1978). J. R. Atkinson, K. J. Brown, and D. Dowson, "The wear of high molecular weight polyethylene, Part 1: The wear of isotropic polyethylene against dry stainless steel in unidirectional motion," Trans. ASME, 100, 208-218 (1978). K. J. Brown, J. R. Atkinson, and D. Dowson, "The wear of high molecular weight polyethylene, Part 2: The effects of reciprocating motion, orientation in the polyethylene, and a preliminary study of the wear of polyethylene against itself," J. Lubricution Technol., 104, 17-22 (1982). D. Dowson and R.T. Harding, "The wear characteristics of UHMWPE against a high density alumina ceramics under wet and dry conditions," Wear, 75, 313-331 (1982). H. C. Amstutz, "Polymers as bearing materials for total hip replacement: A friction and wear analysis," J. Biomed. Muter. Res., 3, 547-568 (1968). D.C. Evans and J. K. Lancaster, "The wear of polymers," Treatise Muter. Sci. Tech., 13, 1979, 85-139.
Received March 1, 1990 Accepted February 5, 1991
counterfaces was 40 to 60% less than that against alumina ceramic counterfaces, and 5 to 10 times lower than with the SUS316L metal counterfaces. Polyethylene wear against metal was more susceptible in saline in which it had 2 to 3 times higher wear rate than with serum. On the other hand, different fluid media had little effect on polyethylene wear against ceramic counterfaces. In each set of tests, the wear factor obtained on an unidirectional wear device showed 10 to 15 times higher values, in comparison to the wear factor estimated on a reciprocating wear device.
INTRODUCTION
All prostheses having bearing surfaces release wear products into joint cavities. The gradual accumulation of these debris particles generates granulomatous reactions in the tissues surrounding the prostheses, and may cause loosening and other problems.1’2High-density polyethylene bearing against metal surfaces normally release variable amounts of plastic and metal due to wear and corrosion. To cope with the problem of wear, Boutin has introduced an alumina/alumina ceramic combination for the socket and ball head in a total hip replacement. Several studies have shown that the wear rate of polyethylene against alumina ceramic is 2 to 3 times less than the wear rate of polyethylene against stainless-steel a l l ~ y . ~ Semlitch -~ et a1.I’ have determined that the wear rate of polyethylene against alumina ceramic is about 20 times lower compared to polyethylene against Co-Cr-Mo alloy. They be‘To whom correspondence should be addressed. Journal of Biomedical Materials Research, Vol. 25, 813-828 (1991) CCC 0021-9304/91/070813-16$4.00 0 1991 John Wiley & Sons, Inc.
KUMAR ET AL.
814
lieve this favorable tribological behavior of ceramic in contact with polyethylene may be due to better corrosion resistance, wettability with liquids, and scratch resistance of the ceramic materials compared to those of metallic implant materials. However, alumina ceramic exhibits a brittle tendency and is sensitive to microstructural flaws." Currently, partially stabilized zirconia ceramic is being recognized for its high strength (and surface finish, making this material potentially suitable for the highly loaded environments found in joint replacement.I2-l4Wagner,I5and o t h e r ~ ' ~have , ' ~ examined the biological reaction of zirconia ceramic in vivo, and have found no difference between its biocompatibility and that of alumina ceramics. However, very little is known about this material's wear resistance properties against UHMWPE. Therefore, the purpose of the present study is to evaluate the wear resistance properties of zirconia ceramic against UHMWPE in association with presently used alumina ceramic/UHMWPE and SUS316L/ UHMWPE combinations. Two principally different wear screening test devices, i.e., unidirectional and reciprocating, are used for this purpose. By using the two different devices, we will try to determine the probable mechanical influence on mechanism of wear, if any.
MATERIALS
Ultra-high-molecular-weight polyethylene (UHMWPE) UHMWPE polymer pins were prepared from Hoechst RCH 1000 surgical grade polyethylene manufactured by the Zeigler low-pressure polymerization process and supplied by Kyocera Corporation, Japan. The molecular weight quoted by manufacturers was in the range of 3.5-4.0 million and a specific gravity of 0.94-0.95. Ceramics and metals counterfaces Mechanical properties of high-density alumina ceramic and yttrium oxidepartially stabilized zirconia (Y-PSZ) ceramic as well as 316L stainless steel (SUS316L) metal alloy are noted in Table I. Y-PSZ ceramic has been manufactured by Kyocera Corporation, Japan, by sintering zirconia powder with 2.53.0 mol.% of yttrium oxide (Y,O,) at a temperature of 1300"-1400"C. Geometrical forms of polyethylene wear pins as well as counterfaces were different in the two wear devices (Fig. 1).Discs used for the unidirectional wear device were annular, having an outer diameter of 60 mm and inner polished diameter of 40 mm with a thickness of 10 mm. For the reciprocating wear device, a rectangular flat surface of 28 mm x 28 mm with a thickness of 10 mm was prepared. It was possible to polish the zirconia ceramic surface of the prosthetic component up to a surface roughness of 0.006 pmRa. For the comparative study, the counterfaces for both devices were polished up to the same surface roughness of about 0.006 pmRa in all three materials, i.e., zirconia ceramic, alumina ceramic, and SUS316L metal.
WEAR OF UHMWPE AGAINST ZIRCONIA
815
TABLE I Selected Mechanical Properties of Zirconia Ceramic, Alumina Ceramic, and SUS 316L Alloy
Properties
Bulk density (lo3 x kg/m2) Flexural strength (MPa) Ultimate tensile strength (MPa) Young’s modulus (GPa) Vickers hardness (MPa) Elongation at fracture (%) Average grain size (pm)
Zirconia Ceramic (Y-PSZ)
Alumina Ceramic
6.0 1000
3.9 500
-
380 17,500 2.5
210 12,500 0.3
SUS 316L Alloy
550 200 1,600 40
Distilled water, physiological saline, and filtered sterile calf bovine serum were chosen as lubricant fluid media. The bovine serum for all sets of tests was from a single lot supplied by CC Laboratories, Cleveland, Ohio, lot number 00846-01 and stored at -20°C until use. Before each test, 30 mg of cefazolin sodium was added in 100 mL of bovine serum to prevent serum degradation by bacteria. MET HODS
Test conditions in the two devices were different, as listed in Table 11. All the polyethylene wear pins used in the devices were 5 mm in nominal thickness and machined from a single block of UHMWPE. For each set of tests, a fresh polyethylene specimen and counterface were used. The tests were repeated three times for zirconia and alumina ceramics and two times for Load
Load
Figure 1. Specimen geometry used in two wear devices: (A) reciprocating wear device, (B) unidirectional wear device.
KUlvlAR ET AL.
816
TABLE I1 Test Conditions in Two Wear Devices Test Conditions
Reciprocating Wear Device
Unidirectional Wear Device
63.58 3.45
12.56 3.00
50
60
Contact Surface Area (mm’) Contact Stress (MPa) Sliding Speed (mm/s.) Temperature (“C) Room Lubricant medium
25
- 26
25
37
- 26 -.
SUS316L metal. Cleaning of the polyethylene pins and counterfaces was done according to the ASTM F732 protocol. Unidirectional wear device (Figure 2) An unsterilized polyethylene cylinder, 4 mm diameter, with a contact surface area of 12.56 mm2 was pressed endwise under a constant axial stress of 3 MPa against the flat surface of the counterface disc. The disc was mounted in a plexiglass chamber having the capacity to contain 40-50 mL of lubricant fluid medium. The wear chamber was driven in a clockwise direction by a variable speed electric motor at a constant speed of 60 mm per second. Wear rate was evaluated by the decrease in polyethylene pin length, and frictional LOADING ARM
U N T E R WEIGHT
LE(ICL4SS CHAMBER MEDIUM
I
> 1
I
c
. I
Figure 2. Outline diagram of unidirectional wear and friction testing machine. Wear rate was estimated by a decrease in polyethylene thickness and frictional force by deformation of the leaf spring through capacitance type displacement probes.
WEAR OF UHMWPE AGAINST ZIRCONIA
817
force by deformation of the leaf spring. Both values were estimated with the help of capacitance type displacement probes (ST-0503, Iwatsu Electric Co., Japan), with gap adjustor (GA-202, Iwatsu Electric Co., Japan). These probes were capable of measuring the displacement with an accuracy of less than 2 pm. These probes were mounted over the machine and connected with transducers (ST-3501, Iwatsu Electric Co., Japan) and a continuous recording chart. Creep deformation and wear volume were recorded without removing the load at any stage. Room temperature of between 24" and 26°C was kept constant throughout the test. Before running the wear chamber, the polyethylene specimen was kept under an initial constant axial stress for a period of 12-24 h or more. During that period, in the first few hours, the polymer deformed rapidly at a rate of about 5 to 10 pm per hour and then at a much slower rate. In the last 3-4 h of loading, when the recording chart showed a steady state, i.e., further creep did not appear on the chart paper, the wear chamber was driven continuously for at least 1 week or more depending on wear prominence.
Reciprocating wear device (Figure 3) Preparation of polyethylene pins and other test conditions used here followed the ASTM F732 protoc01.'~An unsterilized polyethylene cylinder of 9 mm in diameter with a contact surface area of 63.56 mm2was mounted on the wear machine and pressed endwise against the flat counterface surface in a plexiglass wear chamber under constant axial stress of 3.45 MPa. The wear chamber was driven through a 25-mm stroke, or at the sliding distance of 50 mm per cycle with a frequency of 60 cycle per minute. The drive speed, and room temperature of 24"-26"C were kept constant throughout the test. A thermocouple and a heater, connected with a programmable control device,
LOADING ARM
Figure 3. Outline diagram of reciprocating wear and friction testing device. Wear rate was estimated using the same displacement probe used in the unidirectional wear device. Frictional force was measured by strain gauge sensor attached to the leaf spring.
KUMAR ET AL.
818
were inserted into the wear chamber to maintain the temperature of the lubricant f h i d medium at 37°C. Frictional force between the polyethylene specimen and the counterface was monitored by a strain gauge fixed on a leaf spring attached to the transverse bar holding the polyethylene pin. Wear rate was measured by the decrease in wear pin length using the same capacitance type displacement probe used in the unidirectional wear device. To exclude the creep deformation from apparent wear, the polyethylene specimen was treated similar to the unidirectiona1 wear device before running the wear chamber. To prevent dust or other contaminants, both wear devices were fully covered with a transparent polypropylene throughout the test. Compensation for fluid evaporation was made by siphoning distilled water into both devices. Wear factor and coefficient of friction were calculated as follows (Dowson et al."): Wear factor = Wear volume(mm3)/Load(N) x Sliding distance(m) Coefficient of friction = Friction force/Total load Surface changes on polymer pins and counterfaces were analyzed before and after each test under a stereoscopic optical microscope and scanning electron microscope (SEM), as well as visually. Surface topography of both the polyethylene wear pins and the counterfaces was also evaluated on a profilometer before and after each set of tests.
RESULTS
Each set of tests on the reciprocating wear device was allowed to run a total sliding distance of about 65 km or 1.3 million cycles (one cycle = 50 mm) and on the unidirectional wear device for about 30 -40 km continuously. Wear factors for all sets of tests are summarized in Table 111. UHMWPE wear against the SUS316L counterface was 2 to 15 times higher when compared against ceramic counterfaces. Between the two ceramics, the wear factor against zirconia ceramic was about 50% less than the alumina ceramic. The lubricant fluid media had little effect on wear of polyethylene against ceramic counterfaces, but had greater effect against SUS316L metal. With serum lubrication, polyethylene wear against the SUS316L counterface was 2 to 3 times less than in saline. The wear factor obtained in the unidirectional wear device showed a higher trend when compared with the wear factor obtained in the reciprocating wear device. The wear progress of polyethylene as a function of cycles in each set of tests is illustrated in Figure 4. The coefficient of friction between polyethylene and the counterfaces in each set of experiments are summarized in Table IV. An initial high coefficient of friction (0.01 to 0.02 above the mean value during the steady state) against the two ceramic counterfaces was recorded in the first hour of sliding in all three media used. This decreased gradually, stabilized after 2-3 h, and remained close to the values given in Table IV throughout the remainder of
10.7 t 12%
18.2 ? 6% 27.7 ? 30%
3 2
Unidirectional 7.5 t 3% 32.7 2 7 % 90.5 t 40%
1.01 ? 8% 1.81 t 4%
Unidirectional
0.57 3.89
? k
2% 8%
0.45 t 5%
Reciprocating
Saline
0.56 t 14%
Reciprocating
Bovine Serum
3
*Average and range.
Zirconia ceramic Alumina ceramic SUS 316L
Counterfaces
Specimen Number
‘Wear Factor (mm3/N . m) x
TABLE 111 Wear of UHMWPE on Two Different Wear Devices
11.8 ? 4% 37.1 t 10%
8.61 t 11%
Unidirectional
0.68 t 4% 1.12 +. 10%
0.38 -+ 6%
Reciprocating
Distilled Water
F
s o^z
bJ
5e
2
r,
9
m
I CJ
s3
KUMAR ET AL.
820
1.5
I.:\
o.20
vs. alumina c e r a m i c
E 0.25
2. a 2
0.5
0.75
1.0
X106
A
1.25
r 1.5 X106
vs. SUS 316L
2.4-
E
E
2.0
5 0
1.6-
’
1.2,
0.8 0.4 r
0.25
0.5
0.75
1.0
1.25
1.5
X106
Cycles
(C) Figure 4. Wear of UHMWPE as a function of cycles (one cycle = 50 mm), using bovine serum lubrication and a reciprocating wear device. Counterfaces: (A) zirconia ceramic, (B) alumina ceramic, (C) SUS316L.
the test. With the SUS316L counterface, the coefficient of friction in serum lubrication showed a similar trend as with ceramics. However, in saline and distilled water lubricants, the coefficient of friction against SUS316L metal increased (up to 0.1 to 0.3 above the mean value during the steady state) as the test progressed. In the unidirectional wear device, this was accompanied by an increase in lubricant temperature of about 2”-5”C above room temperature. After remaining at a peak for 3-4 h, the coefficient of friction fell gradually to the values shown in Table IV and remained constant for the remainder of the test, except for temporary rises several times during the test. Lubricant temperature also fell and remained steady at about O.Y-1”C above room temperature for the remainder of the test.
0.049 2 8%
0.056 f 10% 0.078 f 22%
3
3 2
Unidirectional 0.082 f 10% 0.115 f 20% 0.156 ? 17%
0.054 f 11% 0.065 f 18%
Unidirectional
0.075 f 2% 0.097 f 8%
0.089 f 10% 0.123 f 5%
5%
0.055
?
Unidrectional
0.044 2 5% 0.061 2 8%
0.028 2 8%
Reciprocating
Distilled Water
0.060 2 3%
Reciprocating
Saline
0.040 f 15%
Reciprocating
Bovine Serum
*Average and range of mean value during steady state.
Zirconia ceramic Alumina ceramic SUS 316L
Counterfaces
Specimen Number
*Coefficient of Friction
TABLE IV Coefficient of Friction of UHMWPE on Two Different Wear Devices
KUMAR ET AL.
822
On visual and stereoscope microscopic examination of surfaces of the two ceramic counterfaces, after the completion of wear tests in three lubricant media, not much difference in surface smoothness compared to their initial surfaces was demonstrated. On the saline- and distilled-water-lubricated SUS316L counterfaces, a pattern of surface scratches running the length of the contact area was seen as the test progressed. By the end of the test there was some evidence of polymer transfer under a stereoscopic microscope (Fig. 5). However, the transfer layer usually was not observed on the serumlubricated SUS316L counterfaces as the metal surface was clean after the test. Profilometer tracings of both ceramic counterfaces as well as SUS316L counterfaces, before and after tests, are shown in Figure 6. In the cases of ce-
Figure 5. Stereoscopic optical microscopic appearance of SUS316L counterfaces against UHMWPE after 1.3 million cycles on a reciprocating wear device: (A) in saline, (B) in distilled water. Counterface sliding direction was horizontal. Before test
in dist.water
In saline
In serum
Zirconia ceramic
Figure 6. Profilometer roughness tracings of the zirconia ceramic, alumina ceramic and SUS316L counterfaces against polyethylene. Tests were performed on a reciprocating wear device. (A, B,C) before test, (U, E, F) in distilled water, (G, H, I) in saline, and (J, K, L) in serum.
WEAR OF UHMWPE AGAINST ZIRCONIA
823
ramic counterfaces, little or no change from their initial surface roughness was recorded after the tests. On the other hand, SUS316L counterfaces became roughened five- to sevenfold from their initial surface roughness in saline and in distilled water, but not in serum lubricant (Fig. 6 F, I, L). The appearance of the worn polyethylene surface, examined with a scanning electron microscope, varied for each of the counterfaces (Fig. 7). Initially the surfaces were covered with ridges formed in the machining process (Fig. 7A). Against zirconia ceramic counterfaces, these machine marks became flattened but still visible over the entire surface indicating that very little wear had occurred (Fig. 7B). Against alumina ceramic counterfaces, the machine tracks were almost invisible in most of the areas and polymer debris adhered to the surface in some places (Fig. 7C). In contrast, after wearing out all machine marks on SUS316L counterfaces a new wear track was established on which islands of redeposited polyethylene were observed (Fig. 7D). The test conditions and duration of test were similar in each set.
DISCUSSION
Wear characteristics of two ceramics and SUS316L metal in association with UHMWPE have been studied. Results presented in this paper showed the advantage of ceramics over a steel counterface, where the wear rate of UHMWPE is decreased by about 5 to 10 times. For the two ceramics, polyethylene wear against the zirconia ceramic is 40 to 50% less than that against the alumina ceramic. Wear estimation
Laboratory screening tests are useful in determining the wear resistant properties of any new prosthetic material prior to clinical use. However, the procedure of measuring accurate wear rate is a controversial subject because of the effect of creep deformation on polyethylene height, and fluid absorption effects on polyethylene weight. McKellop et al.19 and Weightman et aL7 have successfully measured the wear rate by weight loss method. We also attempted to follow the weighing method described in the ASTM F732 protocol with a Sartorius 1-883 weighing machine by weighing the presoak control specimens and the specimens after the completion of tests. The polyethylene specimens were presoaked in serum or water (whichever was used in the wear test) for 10 to 15 days. The steady-state rate of weight gain of presoaked polyethylene specimens was about 2 to 4 pg per day. However, we could not get the accurate wear data (Table V). It is difficult to determine the exact cause of net weight gain presumably due to the greater fluid absorption during the wear test. It may be because of (a) effect of temperature on fluid absorption as the presoak specimen was stored at 4°C to prevent the biodegradation of lubricant medium and the experiment was performed at 37°C; (b) less wear of polyethylene against ceramic counterface which will
824
KUMAR ET AL.
Figure 7. Scanning electron microscopic appearance of UHMWPE surface: (A) initial wear face showed ridges formed in the machining process, (8)on completion of test against zirconia ceramic counterface (original machining marks still visible), (C) on completion of test against alumina ceramic counterface (original machining marks have been almost worn out and surface looks smoother), (D) on completion of test against SUS316L counterface showing original machining marks worn out and formation of a new wear track on polyethylene surface. Note: (B,C, D) after completion of 1.3 million cycles on reciprocating wear device in bovine serum lubrication.
need a prolong sliding distance to give a measurable weight loss; (c) some unknown biochemical and biomechanical mechanism between the polyethylene, counterface and lubricant medium during the wear test which may increase the fluid absorption.
WEAR OF UHMWPE AGAINST ZIRCONIA
825
TABLE V Comparison of Weight Loss of UHMWPE Estimated by Height Loss and Actual Weight Loss
Counterfaces Zirconia ceramic Alumina ceramic SUS 316L alloy
Height Loss Due to Wear (pm/1.3 x lo6 cycles)
Weight Loss Expected Due to Wear (mg/1.3 x lo6 cycles)
Actual Weight Loss Due to Wear (mg/1.3 x lo6 cycles)
14 13 11 20 23 24 42 39
0.836 0.777 0.657 1.195 1.375 1.434 2.510 2.331
-0.059, 0.298 -0.137) 0.388 -0.119, 0.165 0.370 0.836
*( -) Indicate weight gain of polyethylene specimen after completion of tests. Tests were performed on reciprocating wear device, with bovine serum as lubricant.
We found it more accurate and convenient to measure the thickness of polyethylene using a very sensitive displacement probe and gap adjustor capable of measuring with an accuracy of 2 pm. Before sliding, the polyethylene pin was pressed under constant load for a period till the recording chart did not show any further deformation for 3 to 4 h continuously. Thereafter, sliding was started from that point up to the end of the test. And because the decrease in polyethylene pin thickness was recorded continuously without removing the polyethylene specimen or stress during the test, it excluded the possibility of a decrease or increase in the polyethylene pin thickness due to creep deformation during the test, and recovery after the test from the actual wear. However, it was not possible to measure the creep deformation during the sliding which, if any, would continue to decrease.
Effect of lubricant media
The ASTM F732 protocol established bovine serum as an ideal lubricant medium for wear screening tests because of its similarity to normal joint synovial fluid. However, three lubricants were used in this study to determine their effect on polyethylene wear against three counterfaces. Results indicated that saline and distilled water showed different lubrication properties with the SUS316L counterface. In saline, the polyethylene wear against SUS316L metal was two or three times higher compared to bovine serum (Table 111).A transferred polyethylene layer on the SUS316L counterface was only visible when saline or distilled water was used as a lubricant. In bovine serum it was not visible. This may be due to serum proteins acting like a boundary lubricant, as suggested by McKellop” and others. Polyethylene transfer on ceramic counterfaces was not noticed in either fluid medium. Predictably, this may be due to the corrosion resistant properties of ceramic
KUMAR ET AL.
826
counterfaces and a better lubrication mechanism between the ceramic and polyethylene interface.
Two wear devices
In the present study, the wear factor recorded in the unidirectional device was 10 to 15 times higher than the reciprocating wear device results (Table 111). Though the reciprocating device is more relevant to prosthetic joints, the unidirectional device, because of its simplicity, is equally useful for comparative studies of various new and old prosthetic materials in different combinations over a short time span. Further, the severity of wear in the unidirectional device may be due to (a) reduction in the adhesive component of wear with reciprocating motion as suggested by Brown et a1.,2l (b) the ratio of area traversed by the polymer pin on the counterface to the area of the polymer pin, which is about 44 in unidirectional device, and about 4.5 in the reciprocating device. This may influence the wear mechanism in the two devices.
Surface finish It is generally believed, that the wear of UHMWPE decreased against both metal and ceramic counterfaces as the surface finish improved. Weightman7 reported that at a surface finish of about 0.025pmRa, the wear rate of polyethylene against alumina ceramic appeared to be the same as against stainless steel. According to his findings, the talysurf tracing of a metal surface contains more peaks and valleys which might cause an abrasive type of wear. On the contrary, the alumina ceramic surface has flat plateaus and this might be the reason for an adhesive type of wear. In the present study, the profilometric surface analysis did not show much difference among the three counterfaces, i.e., metal and two ceramics. However, SEM examination of zirconia and alumina ceramic surfaces showed great differences in their surface porosities (Fig. 8). This difference in porosity might be one of the reasons for the low wear of polyethylene against zirconia ceramic because, with the test progression, surface defects on the alumina ceramic would have to be filled slowly with transferred polymer.
CONCLUSIONS
(1) This study showed the wear resistant properties of zirconia ceramic against UHMWPE and confirmed the superiority of this ceramic over alumina ceramic and SUS316L metal alloy in terms of low wear and low friction. (2) Different lubricant fluid media had little effect on UHMWPE wear against ceramic counter faces but were prominent against the SUS316L metal. (3) Y-PSZ ceramic may be a biomaterial potentially suitable for low friction arthroplasty because of its better wear resistant properties and high strength.
WEAR OF UHMWPE AGAINST ZIRCONIA
827
Figure 8. SEM micrographs of polished surfaces showing porosity: (A) zirconia ceramic, (B) alumina ceramic. The authors would like to thank Mr. Makinouchi and Mr. Yanagida of Kyocera Co. Ltd., for their kind help in providing the materials and related information.
References 1. U. E. Pazzaglia, L. Ceciliani, M. J. Wilkinson, and C. Dell’Orbo, ”In2. 3.
4. 5.
6.
7. 8.
9.
volvement of metal particles in loosening of metal-plastic total hip prostheses,” Arch. Orthop. Truurnut. Surg., 104, 164-174 (1985). A. Pizzoferrato, L. Savarino, S. Stea, and C. Tarabusi, “Results of histological grading on 100 cases of hip prosthesis failure,” Biomuteriuls, 9, 314-318 (1988). P. Boutin, ”L‘alumina et son utilisation en chirugie de la hanche,” Presse Med., 79, 639 (1971). H. Mittlemeir, ”Selbsthaftende keramik-metal-verbund endoprostheses,” Med. Orthop. Technik, 95, 152-159 (1975). P. Griss and G. Heimke, ”Five years experience with ceramic metal composite hip endoprostheses 1 clinical evaluation,” Arch. Orthop. Truumut. Surg., 98, 157-164 (1981). H. McKellop, I. Clarke, K. Markolf, and H. Amstutz, ”Friction and wear properties of polymer, metal and ceramic prosthetic materials evaluated on a multichannel screening device,” J. Biomed. Muter. Res., 15, 619-653 (1981). B. Weightman and D. Light, “The effect of the surface finish of alumina and stainless steel on the wear rate of UHMW polyethylene,” Biomuterzals, 7, 20-29 (1986). H. Okumura, T. Yamamuro, P. Kumar, T. Nakamura, and M. Oka, “Socket wear in total hip prosthesis with alumina ceramic head,” in Biocerumic, Vol. 1, Proceedings Zst lnternutionul Bioceramic Symposium, H. Oonishi, H. Aoki, and K. Sawai (eds.), 1984, pp. 284-289. K. Kawauchi, Y. Kuroki, S. Saito, H. Ohgiya, S. Sato, S. Kondo, I. Hirose, s. Obara, H. Aso, K. Yamano, s. Sasada, and s. Norose, “Total hip endoprostheses with ceramics head and HDPE socket, clinical wear
KUMAR ET AL.
828
10.
11. 12.
13.
14. 15. 16.
17.
18. 19.
20.
21.
22. 23. 24.
rate," in Orthopaedic Ceramic Implants, Proc. Jap. SOC.Orthop. Ceramic Implants, H. Oonishi and Y. Ooi (eds.), 1984, 4, pp. 253-257. M. Semlitsch, M. Lehmann, and H. Weber, "New prospects for a prolonged functional life-span of artificial hip joints by using the material combination polyethylene/aluminium oxide ceramidmetal," J. Biomed. Muter. Res., 11, 537-552 (1977). J. P. Torre, "Manufacture and engineering properties of ceramics," Proc. Materials Eng. Conf., 5-7th November 1985, London, UK, Mechanical Engineering Publication Ltd., Bury St., Edmund, Suffolk, 1986. K. Shimizu, P. Kumar, M. Oka, Y. Kotoura, Y. Nakayama, T. Yamamuro, T. Yanagida, and K. Makinouchi, "Time dependent change of the mechanical properties of zirconia in vivo," Transaction of the 3rd. World Biomaterial Congress, 21-25th. April, 1988, Kyoto, Japan, Vol. 11, p. 406. K. Makinouchi, T. Yanagida, K. Shimizu, P. Kumar, Y. Kotoura, and M. Oka, "Development of zirconia ceramic for implants," in Orthopaedic Ceramic implants, Proc. Jap. SOC.Orthop. Ceramic Implants, H. Oonishi and Y. Ooi (eds.), 1984, 4. P. Christel, A. Meunier, M. Heller, J. P. Torre, and C. N. IPeille, "Mechanical properties and short term in vivo evaluation of Y-PSZ," J. Biomed. Muter. Res., 23, 45-61 (1989). W. Wagner, "Histologic bone reactions after implantation of alumina and zirconia pins," Trans. 7th European Conf. on Biomaterials, Amsterdam, 1987, p. 44. P. Kumar, K. Shimizu, M. Oka, Y. Kotoura, Y. Nakayama, T. Yamamuro, T. Yanagida, and K. Makinouchi, "Biological reaction of zirconia ceramic," Bioceruinic, Proceedings of 1st. International Bioceramic Symposium, Kyoto, Japan, 1988, Vol. 1, pp. 341-346. ASTM F732-82, Standard practice for reciprocating pin on flat evaluation of friction and wear properties of polymeric materials for use in total joint prosthesis, Annual book of ASTM Standards, ASTM, Philadelphia, 1982. D. Dowson, J. R. Atkinson, and K. J. Brown, "Advances in polymer friction and wear," Polymer Science and Technology Symposia Series, Vol. 58, L.-H. Lee (ed.), Plenum Press, New York, 1974, p. 533. H. McKellop, I.C. Clarke, K.L. Markolf, and H.C. Amstutz, "Wear characteristics of UHMW polyethylene: A method for accurately measuring extremely low wear rates," J. Biomed. Muter. Res., 1!2, 895-927 (1978). J. R. Atkinson, K. J. Brown, and D. Dowson, "The wear of high molecular weight polyethylene, Part 1: The wear of isotropic polyethylene against dry stainless steel in unidirectional motion," Trans. ASME, 100, 208-218 (1978). K. J. Brown, J. R. Atkinson, and D. Dowson, "The wear of high molecular weight polyethylene, Part 2: The effects of reciprocating motion, orientation in the polyethylene, and a preliminary study of the wear of polyethylene against itself," J. Lubricution Technol., 104, 17-22 (1982). D. Dowson and R.T. Harding, "The wear characteristics of UHMWPE against a high density alumina ceramics under wet and dry conditions," Wear, 75, 313-331 (1982). H. C. Amstutz, "Polymers as bearing materials for total hip replacement: A friction and wear analysis," J. Biomed. Muter. Res., 3, 547-568 (1968). D.C. Evans and J. K. Lancaster, "The wear of polymers," Treatise Muter. Sci. Tech., 13, 1979, 85-139.
Received March 1, 1990 Accepted February 5, 1991
