The Journal of Arthroplasty Vol. 14 No. 5 1999

Plast ic i ty-Induced D a m a g e Layer Is a Precursor to Wear in Radiat ion-Cross -Linked UHMWPE

Acetabular C o m p o n e n t s for Total Hip R e p l a c e m e n t

A v r a m A. Edid in , PhD,* Lisa Prui t t , PhD,-t- Cha r l e s W. J e w e t t , MS,$

D e b o r a h J. C rane , M S , t Dan i e l Rober t s , PhD,$ a n d S t e v e n M. Kur tz , P h D §

Abstract: The mechanism for the improved wear resistance of cross-linked ultra-high- molecular-weight polyethylene (UHMWPE) remains unclear. This study investigated the effect of cross-linking achieved by gamma irradiation in nitrogen on the tribologic, mechanical, and morphologic properties of UHMWPE. The goal of this study was to relate UHMWPE properties to the wear mechanism in acetabular- bearing inserts. Wear simulation of acetabular liners was followed by detailed characterization of the mechanical behavior and crystalline morphology at the articulating surface. The wear rate was determined to be directly related to the ductility, toughness, and strain-hardening behavior of the UHMWPE. The concept of a plasticity-induced damage layer is introduced to explain the near-surface orienta- tion of the crystalline lamellae observed in the wear-tested acetabular liners. Cross-linking reduces abrasive wear of acetabular components by substantially reducing--but not el iminating--the plasticity-induced damage layer that precedes abrasive wear. Key words: ultra-high-molecular-weight polyethylene, wear, cross- linking, hip simulator, small punch test, transmission electron microscopy, ultimate properties, morphology, fracture, plasticity-induced damage layer.

Researchers have hypothesized that wear at the articulating surface of total hip replacements is related to the mechanical response of ultra-high- molecular-weight polyethylene (UHMWPE) at large strains under multidirectional loading [1,2]. Direct correlations between the wear behavior of UHMWPE

From *Stryker Howmedica Osteonics, A llendale, New Jersey; -/'Depart- ment of Mechanical Engineering, University of California at Berkeley, Berkeley, California; ~:Exponent Failure Analysis Associates, Menlo Park, California; and §Exponent Failure Analysis Associates, Philadel- phia, Pennsylvania.

Funds were received from a research grant from Stryker Howmedica Osteonics in support of the research material de- scribed in this article.

Submitted March 15, 1998; accepted October 26, 1998. Reprint requests: Avram A. Edidin, PhD, Stryker Howmedica

Osteonics, 59 Route 17, Allendale, NJ 07401. Copyright © 1999 by Churchill Livingstone ~ 0883-540319911405-0016510.0010

and its mechanical behavior at large strains have been limited, however, primarily because bulk ma- terial tests are not performed on specimens scaled for the articulating surface. Advances in miniature specimen testing techniques have greatly facilitated direct mechanical characterization of UHMWPE im- plants [3-7]. The small punch test, originally devel- oped to characterize the tensile and fracture proper- ties of large metallic components using a miniature specimen [8], is particularly well suited for the mechanical characterization of retrieved compo- nents. Small punch testing technology was applied to and validated for virgin UHMWPE using disk- shaped specimens measuring 0.5 mm in thickness and 6.4 mm in diameter [3]. A subsequent investiga- tion demonstrated that the small punch test can be used successfully to characterize as irradiated and

616 P. 0001

Plasticity-Induced Damage Layer Is Precursor to Wear • Edidin et al. 617

oxidatively degraded UHMWPE that had been gamma irradiated in air or nitrogen [4]. Thus, miniature sped- m en testing current ly provides a reliable and effec- tive means to characterize the mechanical proper- ties of UHMWPE using specimens scaled to the articulating surface.

As a semicrystalline polymer, the mechanical behavior of polyethylene is strongly related to its molecular weight and crystalline morphology. For UHMWPE with an average molecular weight of 6 million g/tool, the molecular backbone consists of approximately 400,000 carbon atoms. Visualizing the polymer chain as an entangled strand of string, its unentangled length would be greater than 1 km were the string to have the same aspect ratio as a UHMWPE molecule. Despite the length of the poly- mer chain, UHMWPE in the solid state is a two- phase material, consisting of a crystalline phase embedded within an amorphous phase. The crystal- line phase in UHMWPE consists of folded rows of carbon molecules packed into lamellae, typically 10 to 50 nm in thickness and on the order of 10 to 50 pm in length. The amorphous phase consists of randomly oriented and entangled polymer chains from neighboring molecules. The amorphous phase is also traversed by tie molecules, which intercon- nect remote crystalline domains and provide addi- tional resistance to mechanical deformation. Thus, at a microstructural level, UHMWPE can be consid- ered to be a complex composite material.

Altho.ugh the crystalline phase within UHMWPE is typically randomly oriented within the amor- phous phase, the size, shape, and orientat ion of the lamellae are sensitive to mechanical loading. For example, w h e n deformed in uniaxial tension to strains greater than 0.17, the lamellae wi thin UHMWPE begin to show irreversible changes, which persist even after unloading [9]. The p h e n o m e n o n of irrecoverable, pe rmanen t strain resulting from mechanical loading is termed plasticity, and the study of lamellar plasticity, in particular, holds clues to the mechanical loading history of UHMWPE. Because the liberation of particulate debris from UHMWPE components is expected, at a microscopic level, to involve locally large deformations, from the exami- nation of worn articulating surfaces one should theoretically be able to detect lamellar plasticity, or mechanical damage to the original crystalline mor- phology. This damage would thus be the precursor to wear and wear debris. Only a few detailed studies are currently available in the literature regarding the crystalline morphology of UHMWPE after pro- cessing into ext ruded rod or compression-molded sheet [10,11]. Goldman et al. [12] previously evalu- ated the crystalline structure and morphology of

gamma-irradiated UHMWPE from tibial compo- nents. The near-surface morphology in UHMWPE implants and its likely connect ion to wear mech- anisms, however, has yet to be explicitly investi- gated.

Previous research has suggested that the mechani- cal loading conditions at the articulating surface of acetabular components may be conducive to the development of plasticity-induced damage near the surface [1,2,13-181. While investigating the abra- sion resistance of polyethylene under dry (unlubri- cared) conditions, Pooley and Tabor [13] concluded that molecular orientation was responsible for differ- ences be tween static and kinetic friction at the articulating surface. More recently, wear surface and particle morphology characterization has been performed on retr ieved UHMWPE implants as well as components that have been tested in conven- tional hip joint simulators [ 1,2,17,19,20]. Although wear testing machines do not precisely mimic the complex loading and kinematics of the h u m a n hip joint, conventional biaxial rocking hip joint simula- tors have been shown to reproduce clinically rel- evant wear rates and wear debris for convent ional UHMWPE w h en operated under the appropriate conditions [211. Detailed investigations of the wear surfaces and debris from acetabular components that were previously implanted in patients or exer- cised in joint simulators have suggested that the mechanisms of wear in convent ional UHMWPE are related to orientat ion of long-chain molecules at the articulating surface [ 1,2]. The accumulat ion of plas- tic strains is believed to result in localized anisotropy and strain softening in a direction transverse to the oriented molecular chains [i] . Because of the cross- ing wear paths and multidirectional mot ion preva- lent at the surface of total hip replacements, the shearing of the oriented UHMWPE molecules is hypothesized to be the primary mechanism of abra- sive/adhesive wear in acetabular components [ 1,2l.

Modifications to the structure of UHMWPE, such as cross-linking of the molecular chains, has previ- ously been shown to reduce dramatically abrasive/ adhesive wear in several in vitro hip and knee joint simulator studies [22-24]. It has been suggested that cross-linking of UHMWPE reduces abrasive wear by improving the resistance of the polymer molecules to plastic deformat ion at the articulating surface. The role of cross-linking on the local me- chanical properties and crystalline morphology at the articulating surface, however, has yet to be studied in detail. Consequently, the purpose of the present study was to investigate the effect of cross- linking, as induced by gamma radiation in nitrogen, on the mechanisms of wear in UHMWPE acetabular

P. 0002

618 The Journal ofArthroplasty Vol. 14 No. 5August 1999

componen t s for total hip arthroplasty. This study tested the hypothesis that the wear behavior of UHMWPE in a biaxial rocking hip s imulator is directly related to the mechanical behavior mea- sured at the articulating surface. The small punch test was used to provide quanti tat ive measures of the large de fo rmat ion mechan ica l behav io r of UHMWPE, as characterized by ul t imate strength, ductility, and toughness of the polymer. In addition, using transmission electron microscopy (TEM), we investigated the hypothesis that cross-linking in- creases the abrasive wear resistance of UHMWPE by reducing local plastic deformat ion at the articulating surface.

2,500 N Peak Paul Load Curve (1 Hz)

,urethane Mold

Materials and Methods

Twelve UHMWPE acetabular liners of 2 8 - m m inner d iameter were mach ined f rom a ram-ex- t ruded rod of GUR 4150 HP (PolyHi Solidur, Fort Wayne, IN). The reported molecular weight for the resin was 4.0 to 6.0 million g/tool. Implants were packaged in ni t rogen and submit ted to a commer- cial irradiation facility for repeated doses of g a m m a radiation (Table 1). Sets of 2 implants each were subjected to 0 to 5 repeated doses of g a m m a radia- tion in ni t rogen wi thout violation of the original packaging. Inserts were then subjected to wear testing in a biaxial rocking hip simulator. Wear testing was followed by detailed characterization of the resulting mechanical , physical, and morpho- logic propert ies of the UHMWPE.

Wear Testing

Wear testing was per formed by using an MTS Multi-Station Hip Simulator fitted with commer- cially available coba l t - ch romium femoral heads (MTS Systems Corp., Eden Prairie, MN). Before testing, specimens were presoaked in distilled water for 30 days. The acetabular liners were then m o u n t e d in an anatomic position in po lyure thane molds

Table 1. Summary of Irradiation Cross-Linking and Test Condit ions

Cumulative Average Repeat Gamma Lot Dosage Dosage Density Doses in N2 (No.) (kGy) (kGy) (g/mL)

0 0 0 0.934 1 27.1 27.1 0.938 2 26.3 53.4 0.940 3 26.7 80.1 0.941 4 26.5 107 0.941 5 26.8 133 0.942

~d

Fig. 1. Schematic of the MTS hip simulator used to test acetabular liners after gamma irradiation in nitrogen.

within the testing machine (Fig. 1). The Paul load- ing curve was employed (2500 N maximal load) at a rate of 1 Hz for 3 million cycles.

During testing, triple-filtered bovine calf serum with 0.1% sodium azide, 20 m M ethylenediamine- tetra-acetic acid (EDTA), and 30% distilled water was used as the lubricating medium. Serum was continually replenished with distilled water to ac- count for evaporat ion. The lubricant was replaced every 2.6 × l0 s cycles. The t empera tu re of the lubricant was not controlled during the exper iment but typically ranged be tween 30 ° and 35°C during the test because of frictional heating.

Every 5.7 × 105 cycles, the acetabular liners were r emoved f rom their individual test chambers and cleaned. The mass of each liner was de termined using an analytical balance with a precision of 0.01 mg. Loaded and soaked controls were not used to account for lubricant absorption during the current exper iment . Previous research at the first author ' s institution using loaded and soaked controls has shown that fluid absorpt ion accounts for less than a few percent of the total change in mass of the specimen. Thus, the change in mass for each liner during the present study was not corrected for fluid absorption, which was judged to be negligible. The wear rate was calculated using standard linear regression f rom the steady-state slopes of mass loss as a function of cumulat ive loading cycles over the durat ion of the exper iment .

P. 0003

Plasticity-Induced Damage Layer Is Precursor to Wear • Edidin et al. 619

Top View Schemat ic (Looking Down into Cup)

270 °

Superior (0 °) Multiple Cores Taken Per Acetabular Liner

90 °

Inferior (180 ° )

Lateral Section Schemat ic

i Polar Axis Coring Direction for Small Punch Specimens:

Schematics Not To Scale 45 ° With Respect to Dome

Eig. 2. Schematic of coring locations on the wear-tested acetabular liners. Four cores were taken for the prepara- tion of small punch specimens (SP), and one core was taken for density and morphologic characterization (M).

Small Punch Testing

At the conclusion of wear testing, 4 6 .4 -mm- diameter cores were taken perpendicular to the worn articulating surface of 1 acetabular compo- nen t per irradiation condit ion for the prepara t ion of small punch specimens (Fig. 2). The cores were all taken at 45 ° wi th respect to the polar axis of the cup, and all of the cores were located within the worn area of the articulating surface. To test the hypo th- esis that local variations in loading might influence the measured mechanica l propert ies of the UHM- WPE, the location of each core (ie, at 0 °, 90 °, 180 °, and 270 ° with respect to the super ior- infer ior refer- ence plane) was carefully noted for consideration in the subsequent statistical analysis (Fig. 2).

The surface of each core was carefully mach ined to prepare a 500-1Jm-thick, disk-shaped specimen starting within 25 p m of the articulating surface; a second specimen f rom the core was prepared at a depth of 1.5 to 2.0 m m from the surface. Because of the difficulty in prepar ing specimens near the con- cave surface of the core, however , it was not always possible to manufac tu re all specimens within the required thickness tolerances [3]. As a result, 2 to 5 miniature disk-shaped specimens per cup had accept- ably tight tolerances for subsequent small punch testing. A total of 25 small punch tests were per-

formed on the 6 mult iply irradiated liners; 19 of the tests were pe r fo rmed on specimens starting within 25 p m of the articulating surface, whereas the remaining 6 tests were pe r fo rmed on specimens taken in the depth range of 1.5 to 2.0 m m .

The small punch testing me thod used in this study has been previously described in detail [3,4,8], so only a brief overview of the procedure is provided here. The small punch specimens were placed in a custom-bui l t apparatus and deformed against a hemispher ica l -head punch moving at a constant displacement rate of 0.5 m m / m i n (Fig. 3). The resulting load-displacement curve was character- ized by an initial peak load, an ul t imate load, and ult imate displacement. The work to failure, calcu- lated as the area under the load-displacement curve, provided a measure of toughness. Finally, the shape of the load-displacement curve, which is sensitive to resin, irradiation, and oxidative degradat ion [3,4,8], was interpreted to provide insight into the large- strain plastic deformat ion response of the UHMWPE.

The effects of cross-linking by multiple irradiation doses in ni trogen were statistically evaluated using standard analysis of variance (ANOVA) procedures using SAS Version 6.12 (SAS Institute, Inc., Cary, NC). The effects of secondary exper imenta l design variables, such as specimen location and depth, were also investigated. A P value of .05 was taken to indicate statistical significance.

Density and Morphology

For 1 c o m p o n e n t per irradiation condition, an additional core was taken for the characterizat ion of average density and crystalline morpho logy th rough the thickness. The core was obtained parallel to the

3.8 mm dia. -.-J I---

I / i / /,/°'ell /i/ riI

~n: m thick

Hemispherical Head Punch

Fig. 3. Small punch test apparatus.

P. 0004

620 The Journal of Arthroplasty Vol. 14 No. 5 August 1999

polar axis (Fig. 2), near the center of the worn region at the articulating surface. From each test case, 2 specimens prepared f rom each core were evaluated in the density gradient co lumn prepared with isopropanol and distilled water according to ASTM D1505-85.

The crystalline morpho logy and lamellar align- m e n t of the UHMWPE was de te rmined as a function of depth below the articulating surface using TEM. Using a d iamond knife, representat ive cores were sectioned into 65 -nm slices and placed on grids containing 10 sequential sections spanning a range of 650 n m (0.65 pm) in depth (Fig. 4). The morpho- logic characterization was per formed for each of the 6 dosages by performing TEM on successively ultra- mic ro tomed slices of UHMWPE, starting at the articulating surface and cont inuing to a max imal subsurface depth of 9 pro. Specimens f rom each test group were prepared for TEM by staining the UHMWPE in chlorosulfonic acid (99% concentra- tion) for 6 hours at 60°C. Chlorosulfonation of the UHMWPE stains, cross-links, and stabilizes the amor- phous regions of the material {25]. After acid stain, the samples (allowed to reach ambien t t empera- ture) were rinsed with acetone (0°C) then with distilled water. The samples were dried at 60°C for 1 hour and embedded in epoxy resin wi th a curing t ime of 24 hours at 60°C. Each specimen was u l t ramicro tomed using a d iamond knife into 65-nm- thick sections and collected onto Formvar-suppor ted , carbon-rubid ium slot grids. Slot grids allow for unobst ruc ted viewing over larger areas of the sam- ple. Before TEM evaluation, the specimens were poststained with 0.2% aqueous solution uranyl ac- etate for 3 hours to intensify contrast. Polymer mor- phology was characterized using a JEOL 100CX oper- ating at 80 kV. Lamellar a l ignment was identified in TEM as domains of stacked crystalline lamellae.

Extraction and Swelling Measurements

Extraction and swelling measu remen t s were per- formed to characterize the extent of cross-linking in the same wear- tes ted acetabular liners as were

Samples taken parallel I x I x 5 m m Slaiaed specimen embedded to extracted core in epoxy resin for sectioning

Fig. 4. Schematic of the transmission electron micros- copy sample preparation and slicing orientations taken from the cores from acetabular liners.

subjected to small punch testing, density measure- ment , and morphologic characterization. Specimens of reproducible geomet ry were obtained f rom the 6 wear- tes ted liners by taking 6 .4 -mm-d iame te r cores through the thickness near the equator of the cup. Extraction was pe r fo rmed on 3 specimens per liner using an apparatus similar to that described by ASTM D 2765-95. Each specimen was placed in an individual 500-mL round bo t tom flask at tached to a water-cooled condenser. Into each flask was added p-xylene and I rganox 1010 (0.5% by weight, Ciba- Geigy Corp., Tarrytown, NY), an antioxidant, to reduce degradat ion of the UHMWPE during the prolonged extraction. The UHMWPE specimens, each with a mass of approximate ly 0.3 g, were enclosed in a stainless steel mesh cage (U.S. I20 gage) and suspended within the solvent, which was heated to a rolling boil for 48 hours. Specimens were weighed before and after assembly in the mesh cage using a calibrated analytical balance with a precision of 0.1 mg.

After extraction, specimens were allowed to equili- brate for 2 hours in a fresh bath of p-xylene (also loaded with 0.5 % antioxidant) that was main ta ined at a controlled t empera tu re of 130°C. Specimens were then r emoved f rom the bath; carefully wiped to r emove excess solvent; quickly transferred to the adjacent balance; and then weighed in the hot, swollen, and extracted state. Finally, specimens were dried in a v a c u u m oven at 60°C for 48 hours, which was previously de te rmined to be sufficient t ime for the specimens to reach constant weight. Specimens were weighed once again in the dried and extracted state. Gel content, defined as percent- age of po lymer mass remaining after extraction, was calculated per ASTM D 2765. Swell ratio, defined as the vo lume ratio of the swollen extracted po lymer to the dried extracted polymer, was also calculated in accordance with ASTM D 2765.

Results

Wear Testing

After an initial wear- in period, all of the cups exhibited linear wear behavior for the durat ion of the wear testing (Fig. 5). Five repeated doses of g a m m a irradiation in ni t rogen ( 133 kGy cumula t ive dose) resulted in an 88% reduct ion in average wear rate w h e n compared with the rate after a single irradiation dose (27.1 kGy cumula t ive dose). Based on ANOVA, the cumula t ive dose of g a m m a radia- tion was associated with statistically significant re- ductions in the wear rate ( P < .0001). Multiple irradiation cycles of g a m m a radiation in ni t rogen

P. 0005

Plasticity-Induced Damage Layer Is Precursor to Wear • Edidin et al. 621

250

200

E ~50 fn tn o

..J

100

' ' ' ' I , , ' ' i , , , , i , , , , I ' ' ' ' I . . . .

* 0 Doses (Unirradiated)

• 1 Dose (27.1 kGy Total)

e 2 Doses (53.4 kGy Tota l ) J

----¢---- 3 Doses (80.1 kGy Tota l ) j 1

---.o--- 4 Doses (107 kGy Total)

o---c.--- 5 Doses (133 kGy Total)

50 ...c. ............... c,-----~

~ _ _ _ _ . . . C ~ ~ - - v ..~-----:.--::=~:::=::== :==~:::: -: . . . . . . . . . .

0 ~ - " " ~ ~ ' " - " - ' ' ' ' ' ' ' ,~~ 0.0 0.50 1.0 1.5 2.0 2.5 3.0

Cycles (x 10 G)

Fig. 5. Wear behavior (mass loss) of UHMWPE as a function of loading cycles and cumulative irradiation dose in nitrogen.

• 27.1 kGy 80 o - . . 1 . G y ~-~1 70 . . . . 133 kGy i :

0 il o 40 - 1

30

20

10 : ,

0 i i i L • L ~

0.0 1.0 2.0 3.0 4.0 5.0

Punch D isp lacement (ram)

Fig. 7. Effect of cumulative irradiation dose on load- displacement behavior of UHMWPE during the small punch test.

resulted in progressive and systematic decreases in the rate of wear up to 107 kGy (Fig. 6). Beyond 107 kGy, increased irradiation did not appear to provide fur ther reduct ion in the wear rate.

Small Punch Testing

Mechanical characterizat ion of the wear- tes ted acetabular liners using the small punch test also revealed systematic changes to the large-deforma- tion m e c h a n i c a l b e h a v i o r of the cross- l inked UHMWPE at the articulating surface (Fig. 7). The load-displacement behavior of the irradiat ion-cross- l inked UHMWPE specimens obtained within 25 p m of the articulating surface exhibited an initial peak load, fol lowed by a m e m b r a n e drawing phase dur-

8O

70

% 6 0

E 50

~ 3O

N 2O

• ~ 10

I I I I

L-

, J I , L I , J l , ~ I , L

0 30 60 90 120 Cumulative Sterilization Dose (kGy)

150

Fig. 6. Effect of c u m u l a t i v e i r rad ia t ion dose in n i t rogen on the average s teady-s ta te w e a r rate of UHMWPE.

ing which the UHMWPE deformed in equibiaxial tension a round the head of the punch. The irradia- t ion-cross- l inked UHMWPE exhibited strain-hard- ening behavior after the initial peak load. The strain hardening of the cross-linked UHMWPE was espe- cially notable after 5 cumulat ive doses of g a m m a irradiation in ni t rogen (Fig. 7).

The small punch test reproducibly characterized the large-deformat ion mechanical response of the UHMWPE at the articulating surface for the multi- ply irradiated and wear- tes ted acetabular liners (Table 2). For example, after 1 irradiation dose in nitrogen, the relative exper imenta l uncertainties in initial peak load, ul t imate load, ul t imate displace- ment , and work to failure were each less than 3% (Fig. 8). As indicated by Fig. 8, the small punch test results did not vary considerably as a function of location near the articulating surface.

Cumulat ive radiation doses of greater than 107 kGy were required to induce any measurable differ- ences in the mechanical behavior of UHMWPE near the articulating surface as opposed to into the depth of the acetabular liners. Up to a cumulat ive dose of 80 kGy, the load-displacement behavior of the UHMWPE at a depth of 1.5 to 2.0 m m was qualita- tively similar to the near-surface behavior (Fig. 9A). At cumulat ive doses of 107 and 133 kGy, however , the subsurface mechanical behavior no longer dis- played an initial peak load, whereas the mechanical behavior evaluated at the articulating surface, in contrast, still showed evidence of an initial maxi- m u m in the load-displacement curve (Fig. 9B). The ul t imate load, ult imate displacement, and work to failure did not vary significantly as a function of depth into the liner (P > .05).

Using ANOVA, the wear rate of the acetabular liners was directly correlated to the metrics of the

P. 0006

622 The Journal of Arthroplasty Vol. 14 No. 5 August 1999

T a b l e 2. S u m m a r y o f H i p S i m u l a t o r a n d S m a l l P u n c h Tes t R e s u l t s

Ul t imate C u m u l a t i v e Wear Rate Ini t ial Peak Ul t imate D i sp l acemen t Work to Dosage (kGy) (mg/lO 6 cycles) Load (N) Load (N) (mm) Fai lure (m J)

0 73.0 75.8 69.0 4.34 256 27.1 45.2 73.7 ± 1.0 76.9 -+ 1.7 3.96 ± 0.08 228 ± 8 53.4 15.1 68.5 ÷ 3.6 73.2 -+ 2.9 3.44 -+ 0.09 176 +- 6 80.1 11.1 65.6 + 1.6 74.0 -+ 2.9 3.33 + 0.04 165 -+ 3

107 5.85 64.1 ± 1.8 73.3 -+ 1.8 3.15 _+ 0.07 152 -+ 5 133 5.40 64.8 ± 2.4 69.7 -+ 8.1 3.04 ± 0.13 144 ± 11

load-displacement curve obtained from the small punch tests performed at the articulating surface. The initial peak load was linearly correlated to the wear rate (r 2 = .78, P < .0001) (Fig. 10A). The ultimate load, in contrast, was not significantly affected by the wear rate ( P > .05) (Fig. 10B). Ultimate displacement and work to failure were both found to increase linearly with increasing wear rate (r 2 = .93, P < .0001) (Fig. 10C, D). Thus, im- proved wear performance of radiation-cross-linked acetabular liners in the hip simulator was associated with statistically significant reductions in the initial peak load, ultimate displacement, and work to failure of the UHMWPE determined using the small punch test.

Density and Crystalline Morphology

The average density of the UHMWPE increased nonlinearly with cumulative radiation dose (Fig. 11). The greatest changes in density were observed during the first 27.1 kGy of gamma radiation dos- age, for which density was observed to increase

9 0 , . , i , , , i , , , i , , , - i r , , - - -

S u r f a c e 27 .1 kGy 80 Specimen

70 1 . . . . . Co re2 { . . . . Co re 3

60 ......... co., 4

z so "O

o 40 , - I

30

20

10

0 . . . .

0.0 1.0 2.0 3.0 4.0 5.0 Punch Displacement (ram)

Fig . 8. R e p r o d u c i b i l i t y of t h e s m a l l p u n c h tes t p e r f o r m e d

o n 4 s p e c i m e n s p r e p a r e d f r o m t h e a r t i c u l a t i n g s u r f a c e of

t h e s a m e w e a r - t e s t e d a c e t a b u l a r l iner . B e f o r e w e a r t e s t -

ing, t h e l i ne r w a s g a m m a i r r a d i a t e d in n i t r o g e n w i t h a d o s e of 27 .1 kGy.

from 0.934 g/mL to 0.939 g/mL. For additional cumulative doses, the rate of density change progres- sively decreased. The maximal average density was found to be 0.942 g/mL and occurred at a cumula- tive dosage of 133 kGy.

The TEM analysis showed evidence of lamellar al ignment in all of the wear-tested UHMWPE, re- gardless of cumulative exposure to gamma radia- tion. Previous work [12] and TEM analysis of the

9 0 . . , . . , , . . . , r , , I . . . .

8O

7O

6O

g 50 o 40

. J

a0

20 n

1 0 --o-- Subsurface Specimen

0.0 1.0 2.0 3.0 4.0 50

A Punch Displacement (mm)

9 0 ~ J r r ~ = . [ i ,

8O

7O

60

g s0 ' u

o 40 - I

a0

20 Specimen

1 0 . . . . . Subsurface Specimen 0 i r , i i

0.0 1.0 2.0 3.0 4.0 5 0 B Punch Displacement (ram)

Fig . 9. C o m p a r i s o n of l o a d - d i s p l a c e m e n t b e h a v i o r for

s p e c i m e n s p r e p a r e d f r o m t h e a r t i c u l a t i n g s u r f a c e v e r s u s

s u b s u r f a c e s p e c i m e n s p r e p a r e d at a d e p t h o f 1.5 to 2 m m .

(A) C u m u l a t i v e i r r a d i a t i o n d o s e of 27 .1 kGy. (B) C u m u l a -

t ive d o s e of 133 kGy.

P. 0007

Fig. 10. Correlation of abra- sive wear rate from the hip simulator with (A) initial peak load, (B) ultimate load, (C) ulti- mate displacement, and (D) work to failure as determined by the small punch test. Statis- tically significant correlations were observed between the wear rate and the initial peak load, ultimate displacement, and work to failure (P < .0001).

9 0

80

7O

Z 60

3+0

a.

"~ 30 N = 20

t 0

0

A 5

~" 4 E

-"=2 D

J

0 0

C

Plasticity-Induced Damage Layer Is Precursor to Wear • Edidin et al. 623

~ . : . . . . . . . . . . . . . . I . . . . . . . . . . . . . ;

j___ p • • , L = 10 20 30 40 60 60 70 80 10 20 3 0 4 0 5 0 6 0 7 0 80

Wear Rate (mg/106 cycles) Wear Rate (rag/10 s cycles)

80

9(

8 ¢

7 0

i" 60

,40

30

Z0

10

0

B 30O

26O

. 1 5 0

50

~...8..~ ........................... :" ...........

10 20 30 40 60 60 70 Wear Rate (mgll06 cycles)

0 i i i i

1 0 2 0 3 0 4 0 5 0 6 0

D Wear Rate (mgll0 e cycles)

i

7 0 8 0

bulk mater ial provided evidence that the mic ro tom- ing process did not induce lamellar al ignment, and hence the texture was judged not to be an artifact of the specimen prepara t ion technique. For the virgin, non-cross- l inked UHMWPE, lamellar a l ignment was evident to depths less than 9 p m below the articulat- ing surface (Fig. 12A, B). At a depth of 9 pm, the morpho logy of the virgin UHMWPE was r a n d o m and isotropic (Fig. 12C). The cross-linked UHMWPE also showed evidence of near-surface lamellar align- m e n t after wear testing (Fig. 12D, E). Al though there was still substantial texture deve lopment in the cross-linked UHMWPE, lamellar or ientat ion

0 . 9 5

0 , 9 4 5

. ~ 0 . 9 4 '

0 . 9 3 5

• D o s a g e ( k G y )

0 . 9 3 " " " ' " " " ' " " " ' " " " ' " " " = " : " = " " "

2 0 4 0 6 0 8 0 1 0 0 1 2 0 1 4 0

Dosage ( k G y )

Fig. I l. Average density of the UHMWPE as a function of cumulative radiation dose.

was observed only up to a depth of 4 p m below the articulating surface. This region of lamellar align- m e n t in the wear- tes ted virgin and cross-linked UHMWPE was te rmed the plasticity-induced damage layer.

Extraction and Swelling Measurements

As expected, the unirradiated UHMWPE almost complete ly dissolved after the prolonged extract ion (gel content of 1%). Al though the extract ion and swelling measu remen t s clearly discriminated be- tween irradiated and unirradiated UHMWPE, gel content and swell ratio were found to reach a near ly constant value after a single irradiation dose of 27 kGy in ni trogen (Fig. 13). The average gel content of the ni t rogen-irradiated and wear- tes ted UHMWPE was found to be 99% _+ 1%, whereas the average swell ratio was found to be 2.87 _+ 0.13.

D i s c u s s i o n

The results of this s tudy support the hypothesis that the wear behavior of radiat ion-cross- l inked UHMWPE acetabular componen t s is directly related to the large-deformat ion mechanical behavior deter- mined at the articulating surface. The small punch test has been previously shown to characterize reproducibly UHMWPE in equibiaxial tension to failure. This is the first study, however , in which the

P. 0008

624 The Journal of Arthroplasty Vol. 14 No. 5 August 1999

Fig. 12. After wear testing, the virgin, non-crossqinked (0 kGy) UHMWPE showed evidence of lamellar align- ment at depths of (A) 1.3 to 2.0 lam and (B) 3.3 to 3.9 ~lm from the articulating surface. (C) At a depth of 9 {lm, the morphology of the virgin UHMWPE was random and isotropic. The cross-linked UHMWPE also showed substan- tial lamellar alignment after wear testing but only to a maximal depth of 4 lJm. (D) Depth of 1.3 to 2.0 lam for the cup irradiated with 80 kGy. (E) Depth of 3.3 to 3.9 lJm for the cup irradiated with 133 kGy. The region of lamellar alignment in the wear-tested UHMWPE was termed the plasticity-induced damage layer. (Original magnification × 16,000)

miniature specimen testing technique has been applied at the articulating surface of wear-tested or thopedic implants. In contrast with virgin UHMWPE, which exhibits strain softening after the initial peak load, the radiation-cross-linked materi- als in this study were observed to strain harden.

Previous researchers have shown that radiat ion- cross-linked UHMWPE components exhibit im- proved wear resistance in conventional hip simula- tors [22-24]. It has been proposed that cross-linking

of UHMWPE reduces wear in vitro by improving the resistance of the polymer molecules to lamellar alignment and plastic deformation at the articulat- ing surface. The TEM performed in this study provides evidence that cross-linking does not elimi- nate texture development at the articulating surface but rather reduces the thickness of the plasticity- induced damage layer (Fig. 12). Cross-linking re- duced the plasticity-induced damage layer 2-fold over the non-cross-linked material (as schematically

P. 0009

Plasticity-Induced Damage Layer Is Precursor to Wear • Edidin et al. 625

30 120

25 100

20 80

15 60

1 40 ~

0 ~ 0 0 30 60 90 120 150

Cumulative Sterilization Dose (kGy)

Fig. 13. Effect of cumulative dose on gel content and swell ratio for the UHMWPE.

depicted in Fig. 14). The results of this study suggest that the thickness of this damage layer plays a critical role in the wear mechanism of UHMWPE in acetabular components .

Exposure to high-energy radiation in an inert env i romnent has been shown to promote cross- linking in UHMWPE [26,27]. In the present investi- gation, acetabular liners were subjected to repeated irradiation cycles to reach the desired cumulat ive dosage. Thus, the results of the present study are not directly comparable to prior work because each repeated irradiation cycle can be expected to result in both degradation (ie, chain scission) and cross- linking. The average density of the UHMWPE was observed to increase with repeated irradiation cycles, consistent with chain scission and degradative changes to the polyethylene [28-31]. Despite evi- dence of degradative changes to the UHMWPE during repeated cycles of irradiation, the wear performance improved significantly as a function of cumulat ive dosage. Work by Greer et al. [32] has shown that the improved wear performance of

Plasticity-Induced Damage Layer

(Oriented Lamellae)

Wear Surface .._~'..v_

~(-,. '" Isotropic Potymer , , ' /(I " " " "

A Virgin UHMWPE B Crosslinked UHMWPE

Fig. 14. Schematic of the plasticity-induced damage layer and its relative thickness for (A) the virgin UHMWPE and the (B) radiation cross-linked UHMWPE.

radiation-cross-linked acetabular liners in a conven- tional hip simulator was unaffected by oxidative degradation. Thus, prior research and the present study suggest that the beneficial effects of cross- linking may predominate over the undesirable ef- fects of concomitant degradation.

A no tewor thy finding of this study was that the miniature specimen mechanical testing, density, and morphologic investigations provided more dis- criminating informat ion about the wear- tes ted UHMWPE than did the swelling or extraction experi- ments. Our findings in this regard are consistent with the conclusions of previous researchers [33], who have also noted the lack of disciminatory power of standard swelling techniques to character- ize cross-linking in UHMWPE. Novel and nonstand- ard techniques have been proposed by Muratoglu et al. [33] for direct measurement of swell ratio, which may provide the additional sensitivity that is appar- ently necessary to distinguish be tween UHMWPE of subtly varying cross-link densities.

The complex plasticity mechanisms in polyethyl- ene can be categorized as intralamellar or interlamel- lar [9,34-37]. The intralamellar mechanisms occur within the crystalline lamellae and include twin- ning, martensitic transformation, dislocation chain slip, and unraveling. The interlamellar mechanisms are accommodated through the amorphous region and include interlamellar shear, separation of lamel- lae, tilting or rotation, and cavitation. The findings from the present study suggest that cross-linking affects the depth of crystalline plasticity and sup- presses strain softening in the UHMWPE. Taken together, these observations suggest that radiation cross-linking hinders the interlamellar mechanisms of plastic deformation in the amorphous regions but does not prevent intralamellar plasticity from occur- ring within the crystalline regions of UHMWPE. Thus, the cross-linked UHMWPE investigated in the present study was not immune to abrasive wear but exhibited a significantly lower wear rate because of its smaller plasticity-induced damage layer.

Some of the findings in this study should be interpreted with caution. In addition to improved abrasive wear resistance, radiat ion-induced cross- linking was accompanied by significant decreases in the ductility and toughness of the UHMWPE. The reduction in ductility and toughness for cross-linked UHMWPE is of potential clinical concern, primarily because of their potentially negative impact on the fatigue and f racture resistance of the poly- mer [ 14, 38]. Based on analyses of retrieved acetabu- lar components , it appears that a l though abrasion and adhesion are the pr imary wear mechanisms in the hip, accelerated wear may infrequently be caused

P. 0010

626 The Journal of Arthroplasty Vol. 14 No. 5 August 1999

by scra tches in t he f e m o r a l h e a d or by the i n t r o d u c - t i on of th i rd bodies , such as p o l y ( m e t h y l m e t h a c r y - late) P M M A debr is [2,17,21 ]. Historical ly, no a l ter - a t ions of U H M W P E h a v e b e e n s h o w n to i m p r o v e conc lus ive ly t he clinical, in vivo p e r f o r m a n c e of a h ip a r th rop las ty . Therefore , f r om b o t h a b i o m a t e r i a l s a n d a b i o m e c h a n i c s pe r spec t ive , a n y m o d i f i c a t i o n of U H M W P E for to ta l h ip r e p l a c e m e n t s m u s t necessa r - i ly s t r ike an o p t i m a l b a l a n c e b e t w e e n w e a r per for - m a n c e , ducti l i ty, u l t i m a t e s t reng th , fa t igue e n d u r - ance, a n d f rac tu re res is tance .

Conclusions

In this s tudy, the m e c h a n i c a l a n d m o r p h o l o g i c p r ope r t i e s w e r e d e t e r m i n e d for U H M W P E a c e t a b u - lar l iners tha t w e r e p r e v i o u s l y sub jec t ed to 3 m i l l i on cycles of h ip s i m u l a t o r tes t ing. This w o r k is n o v e l in tha t m i n i a t u r e s p e c i m e n tes t ing m e t h o d s w e r e u sed to d e t e r m i n e the m e c h a n i c a l b e h a v i o r n e a r t he a r t i cu l a t ing surface of t he U H M W P E l iners . Fu r the r , t h e m e c h a n i c a l b e h a v i o r of t h e c r o s s - l i n k e d U H M W P E was c o r r e l a t e d w i t h t he r a d i a t i o n dosage, in vitro w e a r rate, a n d p o l y m e r m o r p h o l o g y .

The fo l lowing conc lus ions can be d r a w n f rom this s tudy:

• E x p o s u r e to h i g h - e n e r g y r a d i a t i o n in an ine r t e n v i r o n m e n t p r o m o t e s c ross - l ink ing a n d signifi- c a n t l y e n h a n c e s a b r a s i v e w e a r r e s i s t a n c e of U H M W P E in a c o n v e n t i o n a l h ip s imula to r .

• The in vitro w e a r b e h a v i o r of r a d i a t i o n - c r o s s - l i n k e d U H M W P E is d i rec t ly r e l a t ed to t he large- d e f o r m a t i o n m e c h a n i c a l p r o p e r t i e s at t he a r t i cu la t - ing surface.

• C ros s - l i nk ing does n o t e l i m i n a t e p las t i c i ty - i n d u c e d t e x t u r e in t he U H M W P E b u t r a t h e r l imits t he t h i ckness of t he p l a s t i c i t y - i n d u c e d d a m a g e l aye r tha t p r e c e d e s ab ras ive wear .

Acknowledgments

Specia l t h a n k s to S. Taylor, R. Windmi l l e r , a n d A. K a l i n o w s k i for t echn ica l ass i s tance a n d for m a n y he lp fu l d iscuss ions .

References

1. Wang A, Sun DC, Yau S-S, et ah Orientation softening in the deformation and wear of ultra-high molecular weight polyethylene. Wear 203:230, 1997

2. Jasty M, Goetz DD, Bragdon CR, et al: Wear of polyethylene acetabular components in total hip ar- throplasty: an analysis of one hundred and twenty-

eight components retrieved at autopsy or revision operations. J Bone Joint Surg Am 79:349, 1997

3. Kurtz SM, Foulds JR, Jewett CW, et al: Validation of a small punch testing technique to characterize the mechanical behavior of ultra-high molecular weight polyethylene. Biomaterials 18:1659, 1997

4. Kurtz SM, Jewett CW, Foulds JR, Edidin AA: A miniature-specimen mechanical testing technique scaled to the articulating surface of polyethylene components for total joint arthroplasty. J Biomed Mater Res (Appl Biomater) 1998 48:75, 1999

5. White SE, Paxson RD, Tanner MG, Whiteside LA: Effects of sterilization on wear in total knee arthro- plasty. Clin Orthop 331:164, 1996

6. Sutula LC, Collier JP, Saum KA, et al: Impact of gamma sterilization on clinical performance of poly- ethylene in the hip. Clin Orthop 319:28, 1995

7. Collier JP, Sperling DK, Currier JH, et ah Impact of gamma sterilization on clinical performance of poly- ethylene in the knee. J Arthroplasty 11:377, 1996

8. Foulds JR, Woytowitz P J, Parnell TK, Jewett CW: Fracture toughness by small punch testing. J Testing Eval 23:3, 1995

9. Butler ME Donald AM, Ryan A J: Time resolved simulataneous small- and wide-angle X-ray scatter- ing during polyethylene deformation: II. Cold draw- ing of linear polyethylene. Polymer 39:39, 1998

10. Bellare A, Cohen RE: Morphology of rod stock and compression-moulded sheets of ultra-high-molecular- weight polyethylene used in orthopaedic implants. Biomaterials 17:2325, 1996

11. Pruitt L, Bailey L: Factors affecting the near-threshold fatigue behavior of surgical grade ultra high molecu- lar weight polyethylene. Polymer 39:1545, 1998

12. Goldman M, Gronsky R, Ranganathan R, Pruitt L: The effects of gamma radiation sterilization and age- ing on the structure and morphology of medical grade ultra-high molecular weight polyethylene. Polymer 37:2909, 1996

13. Pooley CM, Tabor D: Friction and molecular struc- ture: the behavior of some thermoplastics. Proc Roy Soc Lond 329:25i, 1972

14. Wang A, Stark C, Dumbleton JH: Role of cyclic plastic deformation in the wear of UHMWPE acetabular cups. J Biomed Mater Res 29:619, 1995

15. Wang A, Essner A, Stark C, Dumbleton JH: Compari- son of the size and morphology of UHMWPE wear debris produced by a hip joint simulator under serum and water lubricated conditions. Biomaterials 17:865, 1996

16. Wang A, Essner A, Polineni VK, et ah Wear mecha- nisms and wear testing of ultra-high molecular weight polyethylene in total joint replacements. Presented at the Polyethylene Wear in Orthopaedic Implants Work- shop, Society for Biomaterials, New Orleans, April 30-May 5, 1997

17. McKellop HA, Campbell P, Park SH, et ah The origin of submicron polyethylene wear debris in total hip arthroplasty. Clin Orthop 311:3, 1995

P. 0011

Plasticity-Induced Damage Layer Is Precursor to Wear • Edidin et al. 627

18. Cooper JR, Dowson D, Fisher J: Macroscopic and microscopic wear mechanisms in ultra-high molecu- lar weight polyethylene. Wear 162-164:378, 1993

19. Wang A, Stark C, Dumbleton JH: Mechanistic and morphological origins of ultra-high molecular weight polyethylene wear debris in total joint prostheses. Proc Inst Mech Eng 210:i41, 1996

20. Bragdon CR, O'Connor DO, Lowenstein JD, et al: The importance of multidirectional motion on the wear of polyethylene. Proc Inst Mech Eng 210:157, 1996

21. Sauer WL, Anthony ME: Predicting the clinical wear performance of orthopaedic bearing surfaces. In Ja- cobs J J, Craig TL (eds): Alternative bearing surfaces in total joint replacement. American Society for Testing and Materials; West Conshohoken, PA, 1998

22. McKellop HA, Shen FW, Yu Y J, et al: Effect of sterilization method and other modifications on the wear resistance of UHMWPE acetabular cups. Pre- sented at the Polyethylene Wear in Orthopaedic hnplants Workshop, Society for Biomaterials, New Orleans, April 30-May 5, 1997

23. Jasty M, Bragdon CR, O'Connor DO, et al: Marked improvement in the wear resistance of a new form of UHMWPE in a physiologic hip simulator. Presented at the 43rd Annual Meeting of the Orthopaedic Re- search Society, San Fransisco, February 9-13, 1997

24. Wang A, Polineni VK, Essner A, et al: Effect of radiation dosage on the wear of stabilized UHMWPE evaluated by kip and knee joint simulators. Presented at the 23rd Annual Meeting of the Society for Bioma- terials, San Fransisco, March 18-22, 1997

25. Sawyer LC, Grubb DT: Polymer microscopy. Chap- man & Hall, New York, 1987

26. Streicher RM: Ionizing irradiation for sterilization and modification of high molecular weight polyethyl- enes. Plast Rubber Proc Appl 10:221, 1988

27. Rose RM, Goldfarb EV, Ellis E, Crugnola AN: Radia- tion sterilization and the wear rate of polyethylene. J Orthop Res 2:393, 1984

28. Rimnac CM, Klein RW, Betts F, Wright TM: Post- irradiation aging of ultra-high molecular weight poly- ethylene. J Bone Joint Surg Am 76:1052, 1994

29. Kurtz SM, Rimnac CM, Bartel DL: Degradation rate of ultra-high molecular weight polyethylene. J Orthop Res 15:57, 1997

30. Bostrom MP, Bennett AP, Rimnac CM, Wright TM: The natural history of ultra high molecular weight polyethylene. Clin Orthop 309:20, 1994

31. Kurtz SM, Rimnac CM, Santner TJ, Bartel DL: Expo- nential model for the tensile true stress-strain behav- ior of as-irradiated and oxidatively degraded ultra high molecular weight polyethylene. J Orthop Res 14:755, 1996

32. Greer KW, Schmidt MB, Hamilton JV: The hip simula- tor wear of gamma-vacuum, gamma-air, and ethyl- ene oxide sterilized UHMWPE following a severe oxidative challenge. Presented at the 44th Annual Meeting of the Orthopaedic Research Society, New Orleans, February 21-24, 1998

33. Muratoglu OK, Cook JL, Jasty M, Harris WH: A novel technique to measure the cross-link density of irradi- ated UHMWPE. Presented at the 44th Annual Meet- ing of the Orthopaedic Research Society, New Or- leans, February 21-24, 1998

34. Lin L, Argon AS: Review: structure and plastic defor- mation of polyethylene. J Mater Sci 29:294, 1994

35. Bowden PB, Young R J: Review: deformation mecha- nisms in crystalline polymers. J Mater Sci 9:2034, 1974

36. Keller A, Pope DP: Identification of structural pro- cesses in deformation of oriented polyethylene. J Mater Sci 6:453, 1971

37. G'Sell C, Dahoun A: Evolution of microstructure in semi-crystalline polymers under large plastic deforma- tions. Mater Sci Eng 175:183, 1994

38. Gul R: Improved UHMWPE for use in total joint replacement. Dissertation, Massachusetts Institute of Technology, Boston, September 1997

P. 0012