Journal of Applied Biomaterials & Biomechanics 2004; 2: 20-28

Influence of heat treatment on structural,

mechanical and wear properties of crosslinked
UHMWPE
R. CHIESA1, M. MOSCATELLI1, C. GIORDANO1, F. SICCARDI2, A. CIGADA1
Department of Chemistry, Materials and Chemical Engineering G.Natta, Politecnico di Milano, Milano - Italy
Medacta International SA, Castel S. Pietro - Switzerland

1
2

ABSTRACT: New crosslinked ultra high molecular weight polyethylenes (UHMWPEs) have recently been developed, characterized and introduced in clinical applications. UHMWPE cross-linking treatments are very promising for reducing osteolysis induced by wear debris. The irradiation type, gamma or beta, the dosage and the thermal treatment performed during or following the irradiation process are all factors affecting polyethylene wear resistance. Thermal stabilization treatments performed after or during the irradiation process at a temperature above melting point (i.e. >130 C) have been proven to effectively remove
the free radicals generated during irradiation from UHMWPE, but their effect on the mechanical properties of UHMWPE are
not completely clear. In addition to wear rate reduction, maintaining good mechanical properties is fundamental aspect in designing the new generation of crosslinked UHMWPE for artificial load bearing materials, especially considering the application in total knee replacements.
In this study, we investigated the influence of different stabilization treatments, performed after gamma irradiation, on structural, wear and mechanical properties of UHMWPE. We performed four different stabilization treatments, with different temperatures and cooling rates, on 100 kGy gamma irradiated UHMWPE. Structural properties of UHMWPE were assessed by
differential scanning calorimetry (DSC). To assess the mechanical performance of the materials, uni-axial tensile tests were performed according to the ASTM D638 standard, bi-axial tension performance was evaluated by small punch tests (ASTM
F2183-02), toughness resistance was evaluated by the Izod method (ASTM F648), and cold flow resistance was analysed by a
dynamic compressive test. Evaluation of wear resistance was by a multidirectional pin-on-disk screening machine. Materials
considered were in aged and non-aged conditions. Results confirmed that cross-linking greatly enhances UHMWPE wear
resistance, but introduces some detrimental effects on the mechanical properties. In this study, we found that the negative effects on the mechanical properties of crosslinked UHMWPE can be modulated, to some extent, by choosing a thermal stabilization treatment at a correct temperature and cooling rate. (Journal of Applied Biomaterials & Biomechanics 2004; 2: 20-8)
KEY WORDS: UHMWPE, Cross-linking, Wear, Thermal treatment
Received 27/12/03; Revised 14/01/04; Accepted 16/01/04

INTRODUCTION
In the last few decades, ultra high molecular weight
polyethylene (UHMWPE) has been the chosen material for load bearing surfaces in total joint replacements. Excellent tribological properties and
good mechanical performance established its success and worldwide distribution. Nevertheless, it
has become evident that in the long term, the loosening and failure of joint prostheses is closely asso-

ciated with osteolysis induced by polyethylene wear

debris (1). Currently, UHMWPE wear, often associated with its oxidation and degradation, is recognized as the most important limiting factor compromising the long-term performance of acetabular components in total hip replacements (2, 3).
Polyethylene cross-linking is known to improve
abrasion resistance for industrial applications (4).
Since the late 1990s, studies confirmed that
UHMWPE cross-linking for orthopedic artificial

1722-6899/020-09$15.00/0

Societ Italiana Biomateriali

Chiesa et al

joints generates a wear and oxidation-resistant load

bearing material, capable of potential dramatic improvements in the clinical performance of total
joint replacements (5-13). Many cross-linking treatments have been developed and clinically applied
to total hip replacements. Currently, > 50% of
UHMWPE used for hip replacements in the USA is
crosslinked. However, in order to extend the application safely of cross-linked UHMWPE to highly
mechanically stressed joints, such as the knee, there
is a need for a better understanding of the relationship between cross-linking treatments, wear
and mechanical performance.
This study aimed to understand the role of thermal
stabilization treatment on the properties of
crosslinked polyethylene.

centage of crystallinity was determined by assuming

a value of 289 J/g for the fusion heat of perfect
crystalline polyethylene (14).

Tensile test
Before and after accelerated ageing the irradiated
UHMWPE and the non-irradiated control underwent
tensile mechanical testing according to ASTM D638.
The tensile properties of the samples were evaluated at
room temperature using an Instron universal tester
(model 4302) at a cross-head speed of 50 mm/min.
The specimens, machined from the bar, were subsequently cut with a sample die and preconditioned for
24 hr at 23 1 C with a 50 5% UR. Yield stress, ultimate tensile stress and strain were recorded. Five specimens were tested for each material.

MATERIALS AND METHODS

Toughness resistance

Medical grade UHMWPE resin GUR 1020, originally produced by Ticona, and compression molded by
a commercial converter (Poly Hi Solidur, Germany) was used for all experiments.
UHMWPE bars were crosslinked using 100 kGy of
rays and were subsequently thermally stabilized above
melting temperature to eliminate free radicals.
Table I summarizes the materials included.
The materials were artificially aged according to
the ASTM F2003-00 method B for 14 days at 5 bar
of pure O2 at 70 C (oxygen bomb ageing).

Impact resistance strength was assessed with an

Izod Ceast Resilimpactor machine according to
ASTM F648 (Annex A1), using double notched
specimens. Each material was tested 6x. Aged and
non-aged specimens were tested.

Differential scanning calorimetry

Crystallinity degree and material melting temperatures were determined by differential scanning
calorimetry (DSC) (Seiko 6200 SII, software Extar
6000). Indium was used for temperature calibration
and heat of fusion.
Heating cycles were run from 25-200 C at a rate of
10 C/min in nitrogen. The weight of the specimens was in the range of 2.58 0.14 mg. The per-

Small punch test

Bi-axial tension tests were performed on materials
using the small punch test method, following ASTM
F2183-02, before and after accelerated ageing.
The specimens, machined from the bar, were cut
with a sample die into disk-shapes (thickness 0.5
mm, diameter 6.4 mm), and then placed in the
small punch testing apparatus. A hemisphericalhead punch applied stress, moving at a constant displacement rate of 0.5 mm/min.
From the load-displacement curve obtained by the
small punch test, the initial peak load, the ultimate load and the ultimate displacement were calculated. The energy to failure, calculated by integrating the load-displacement curve, was assumed

TABLE I - MATERIALS

NT
25 kGy
130 W
130 1/10
155 W
155 1/10

Irradiation
dose (kGy)

Temperature
and duration of thermal
treatment (C)

Cooling rate

25
100
100
100
100

130 C - 72 hr
130 C - 72 hr
155 C - 72 hr
155 C - 72 hr

fast
slow (0.1 C/min)
fast
slow (0.1 C/min)

21

Influence of heat treatment on structural, mechanical and wear properties of crosslinked UHMWPE

as a measure of toughness. It was repeated 5x for

each material. Scanning electron microscopy
(SEM) evaluation was performed to better characterize differences in specimen morphology and
fracture surface characteristics. SEM analysis was
performed on selected samples after small punch
testing. Specimens were gold-sputtered and examined using a scanning electron microscope (Stereoscan S360, Cambridge Instruments), operating at
10 keV, at magnifications of 15x (specimens view)
and 200x (fracture surface analysis).
A double-tailed Students t-test (Statview 5.0.1, SAS
Institute, Inc) was used to assess the statistical significance of differences in measurements.

Dynamic compressive test

All non-aged materials underwent dynamic compressive testing to assess their cold flow. Cylindrical
specimens, shaped according to ASTM D695 (diameter = 15 mm, h = 30 mm), were machined from
the treated bars.
Tests were performed using an Instron mod. 8062
machine. Figure 1 shows the testing apparatus. A
compressive sinusoidal load was applied (minimum
stress 0.45 MPa, maximum stress 4.5 MPa) for
250,000 cycles, at a frequency of 1 Hz. This particular load and frequency were chosen to simulate
stresses during a walking cycle (15).
Specimens were kept in Ringer solution at a temperature of 37 C during the entire testing. Four
specimens were tested for each material. The displacement was measured using an Instron extensometer with a resolution of 0.1 mm. During the
test, the highest and lowest displacement values of
each cycle were recorded. This displacement was
normalized by the specimen height to obtain a
measure of strain. Finally, the instant cold flow was
calculated by subtracting the offset recorded with
each cycle from the 1st cycle of the test, assumed to
be the elastic deformation.
The maximum and minimum strain considered at
each cycle i were, respectively:

Fig. 1 - Cold flow testing apparatus.

1. Cylindrical housing (PMMA); 2. Stainless steel pushing
punch; 3. Connection flange; 4. Extensometer connecting rod; 5.
Connection flange; 6. Extensometer; 7. Water heating system; 8.
Covering.

stant cold flow deformation.

To assess the cold flow deformation (creep), the resulting data curve was interpolated with an allometric model (16):

cold flow = a .N b
where:
N is the number of cycles;
a and b are the characteristic parameters of cold
flow, evaluated by the iterative Levenberg-Marquardt algorithm.
In a double logarithmic chart, parameter a is the
intersection with the deformation axis at initial cycles, and indicates the initial deformation. Parameter b is the slope of the line indicating the cold flow
deformation rate.
Finally, the average strain standard deviation
(SD) at 250,000 cycles was calculated for each material.

Wear test
where:
l0 is the initial height in millimeters of the cylindrical specimen;
S is the displacement in micrometers recorded by
the extensometer at cycle i;
is the cold flow deformation at cycle i.
The average strain value was calculated for each cycle, and the value obtained was assumed as the ini

22

Wear testing was carried out using a six-station pinon-disk machine (Ortho-POD, AMTI Watertown,
MA, USA) applying a multidirectional sliding motion. Highly polished CoCrMo disks (Ra < 0.05)
were used as counter-specimens.
Calf bovine serum (Sigma B9433) diluted at 25%
volume concentration in de-ionized water was used
as a lubricating medium. A figure-of-eight shaped

Chiesa et al

loading curve and a variable loading simulating a

Paul curve were applied (15). Load ranged from 04.5 MPa with a minimum of 1.7 MPa and a maximum of 3.4 MPa.
The temperature of the lubricant was uncontrolled during the experiment, but it typically
ranged from 25-30 C. Three pins for each material were tested up to 500,000 cycles. Every 125,000
cycles, the pins were removed from the pin-ondisk machine and cleaned in accordance with
ASTM F2025-00.
UHMWPE pin wear was determined by a gravimetric
method using an analytical balance with a precision
of 0.1 mg. Load-soak controls were used to account
for lubricant absorption during the experiment.

ture, slow cooled samples showed higher Tm and

cristallinity than quenched samples.
The increase in crystallinity observed by DSC
analysis after irradiation could be because of an
improvement in crystal perfection. It has been
suggested (17-19) that irradiation-induced chain
scission permits the broken polymer chains in the
amorphous regions of the polymer to fold and recrystallize, as well as allowing an increased perfection of existing folded-chain crystallites.
The low crystallinity of quenched samples indicated
that rapid cooling from melting can introduce
some crystal imperfection, preventing higher crystallinity development.
In particular, this was seen in irradiated samples
where crosslinks can inhibit re-crystallization.

RESULTS AND DISCUSSION

Tensile test and toughness resistance

Differential scanning calorimetry

Tensile and toughness tests were performed on all

materials before ageing and on non-treated (NT),
25 kGy and 155 1/10 materials after artificial ageing. The results of tensile and impact toughness
measured by IZOD testing are summarized in Table
III (non-aged materials) and Table IV (aged materials), and in Figures 2, 3 and 4. Results are displayed as mean value SD.

Table II shows the thermal properties assessed by

DSC analysis of the different UHMWPEs.
After gamma irradiation at 100 kGy, the samples exhibited an increase in crystallinity degree and melting temperature. In addition, for thermal stabilization treatments performed at the same tempera-

TABLE II - PEAK MELTING TEMPERATURE, HEAT OF FUSION AND DEGREE OF CRYSTALLINITY

Materials

Tm (C)

h (J/g)

Cryst %

NT
25 kGy
130 W
130 1/10
155 W
155 1/10

135.8
136.1
135.8
136.8
134.7
137.8

104.0
151.7
110.0
126.7
115.1
119.5

36.0
52.5
38.0
43.7
39.7
41.2

TABLE III - TENSILE PROPERTIES AND TOUGHNESS (IZOD) RESULTS

Materials

Yield strength
(MPa)

Strain at break

UTS
(MPa)

Impact strength
(kJ/m2)

NT
25 kGy
130 W
130 1/10
155 W
155 1/10

23.06 2.29
22.77 0.58
21.87 0.29
21.91 0.16
18.47 0.38
19.85 0.62

401.7 18.1
462.8 15.4
236.0 10.8
246.0 6.60
261.6 9.20
245.3 12.2

45.86 5.16
52.98 3.83
40.19 2.24
44.32 1.64
33.20 2.34
36.35 2.97

156.3 7.3
123.3 4.1
060.9 3.5
060.9 2.4
069.6 4.2
070.2 4.5

23

Influence of heat treatment on structural, mechanical and wear properties of crosslinked UHMWPE

Fig. 2 - Effect of accelerated ageing on ultimate tensile strength.

Irradiation had little effect on yield stress measured

by tensile testing. Strain at break reduced by 35%
for all crosslinked materials. There was also a reduction in strength. Irradiated material stabilized
at 155 C, showed a slightly reduced ultimate tensile strength than material stabilized at 130 C.
Another relevant reduction in mechanical properties associated to cross-linking treatment was seen
in impact strength measured by IZOD testing. In
comparison to NT, all crosslinked materials showed
a reduction in impact resistance strength ranging
from 55-60%. Among the crosslinked materials, the
thermal stabilization treatment at 155 C showed a
slight increase (+5%) in toughness resistance (ttest, p < 0.01).
Tensile tests performed on crosslinked 155 1/10
material did not show a considerable difference in
the tensile properties pre- vs. post-artificial ageing.
On the contrary, a reduction in ultimate tensile
strength was seen on both NT and 25 kGy materials,

38% and
83% respectively. Considering strain
at break, 25 kGy material showed an important reduction (
78%), while NT and 155 1/10 materials
did not vary.
After ageing, the impact resistance strength decreased in all materials. There was a difference in
toughness resistance for 25 kGy air sterilized material (
81% after ageing), while NT and 155 1/10

Fig. 3 - Effect of accelerated ageing on strain at break.

Fig. 4 - Effect of accelerated ageing on impact resistance strength.

materials showed little decrease (
18% and
24%,

respectively).

Small punch test

Table V (non-aged materials) and Table VI (aged
materials) summarize the results of small punch
tests. Values are expressed as mean SD.
The NT and the sterilized specimens displayed similar small punch behavior, characterized by an initial peak load recorded during initial bending of
the disk-shaped specimen, followed by a drawing
phase. Compared with the control materials, the

TABLE IV - TENSILE PROPERTIES AND TOUGHNESS (IZOD) RESULTS AFTER ACCELERATED AGEING
Materials

Yield strength

Strain at break

(MPa)
NT
25 kGy
155 1/10

24

23.36 0.16
28.13 0.69
20.83 0.39

381.4 31.6
102.2 65.9
256.6 23.2

UTS

Impact strength

(MPa)

(kJ/m2)

28.47 1.24
08.83 4.48
31.81 4.47

128.4 7.2
023.3 7.4
053.3 1.9

Chiesa et al

TABLE V - SMALL PUNCH TESTS RESULTS

Materials

Peak load (N)

Ultimate displacement (mm)

Ultimate load (N)

Work to failure (mJ)

NT
25 kGy
130 W
130 1/10
155 W
155 1/10

69.8 1.6
68.3 3.1

4.30 0.10
4.16 0.07
3.19 0.33
3.04 0.07
3.50 0.25
3.34 0.25

54.5 2.4
53.9 4.0
84.5 7.6
85.4 1.5
71.0 4.2
86.0 5.2

245 12
232 7
190 28
189 5
168 18
201 24

TABLE VI - SMALL PUNCH TESTS RESULTS AFTER ACCELERATED AGEING

Materials

Peak load (N)

Ultimate displacement (mm)

Ultimate load (N)

Work to failure (mJ)

NT
25 kGy
130 W
130 1/10
155 W
155 1/10

73.4 2.7
72.7 2.5
78.5 0.9
76.8 1.3

4.17 0.13
4.10 0.20
2.74 0.06
2.89 0.19
3.11 0.52
3.08 0.27

56.8 2.4
52.4 2.0
73.5 3.1
70.8 5.4
081.7 14.2
083.6 11.2

247 11
240 70
186 40
184 11
183 48
184 27

Fig. 5 - Representative curves of load-displacement of UHMWPE

untreated (a) and crosslinked (b).

crosslinked materials behaved differently, characterized by an ultimate load increase, an ultimate

displacement reduction, and the absence of the
ductile drawing phase in the load-displacement
curve (Fig. 5).
The small punch curve of crosslinked UHMWPE
did not show any drawing phase. This indicated the
relatively brittle behavior of UHMWPE after crosslinking.
After ageing, the mechanical response of NT, 25

kGy and materials stabilized at 155 C did not

change significantly (t-test, p > 0.5). Both materials
treated at 130 C showed a reduction in ultimate
load and ultimate displacement (t-test, p < 0.01).
SEM analysis performed on small punch tested
specimens proved that cross-linking treatment was
associated with a reduction in ductility.
The SEM images of NT and 25 kGy air sterilized
specimens showed large-scale plastic deformation
and reduced apex thickness (Fig. 6 a and 6 b). Ductile tearing and rupture characterize the fracture
surfaces.
The crosslinked materials showed very little plastic
deformation with a typical brittle surface morphology (Fig. 7 a and 7 b).
Although machining lines were evident in specimens,
they did not affect the outcome of the test results.

Dynamic compressive tests

Tables VII, VIII and Figure 8 summarize the results
of the dynamic compressive tests.
130 W and 130 1/10 materials showed the lowest
strain at a cyclic loading of 250,000. They also
showed a low initial deformation, but an increased
strain with a slightly higher ratio (see parameter a
and b values in Table VII and Figure 8).
25

Influence of heat treatment on structural, mechanical and wear properties of crosslinked UHMWPE
Fig. 6 - a) Representative scanning electron micrographs of
UHMWPE small
punch test specimens
of untreated or sterilized materials after
testing to failure (original magnification
15x); b) fracture surface (original magnification 200x).

Fig. 7 - a) Representative scanning electron micrographs of

cross-linked UHMWPE small punch test
specimens after testing to failure (original magnification
15x); b) fracture
surface (original magnification 200x).

TABLE VII - DYNAMIC COMPRESSIVE TESTS: ALLOMETRIC MODEL PARAMETERS

a
b

NT

25 KGy

130 1/10

130W

155 1/10

155W

2542

2628
0.09609

1639
0.11236

1321
0.13049

2872
0.08649

2808
0.10068

Wear test
Table IX and Figure 9 summarize the effect of gamma irradiation and stabilization treatment on
UHMWPE wear behavior.
Before ageing, crosslinked and stabilized materials
exhibited an improved resistance to wear over the
NT and 25 kGy sterilized materials. The mean
weight loss measured for the crosslinked specimens
was nearly zero, regardless of the thermal treatment performed (temperature and cooling
rate).
After ageing, with the crosslinked materials, there
was no difference in the amount or wear rate. On
26

TABLE VIII - DYNAMIC COMPRESSIVE TESTS: AVERAGE STRAIN AND SD AT 250,000 CYCLES
Materials

Average strain at 250,000 cycles

(mm/mm 106)

NT

8650 510

25 Kgy

8370 480

130 W

6470 950

130 1/10

6460 1120

155 W

9670 460

155 1/10

8120 1090

Chiesa et al

Fig. 8 - Allometric model in double logarithmic scale of all considered materials.

Fig. 9 - Effect of accelerated ageing on weight loss at 500,000

cycles.

TABLE IX - WEAR (mg) AT 500,000 CYCLES: MEAN SD

variations for 25 kGy gamma sterilized materials,

while 155 1/10 materials showed negligible differences in all properties considered. Finally,
cold flow did not seem affected by cross-linking
treatments.

Materials

Wear
(mg)

Wear after
aging (mg)

NT

3.5 0.4

0.4.3 2.3

25 kGy

5.4 0.5

.33.2 9.3

130 W

0.1 0.40.

0.8 0.9

130 1/10

0.3 0.60.

0.8 0.8

155 W

0.2 0.4

0.0.9 3.6

155 1/10

0.0 0.4

0.0.2 0.3

the contrary, the 25 kGy air sterilized polyethylene

was significantly affected by the ageing process,
and the weight loss of 25 kGy air sterilized aged
material was significantly increased compared to
aged NT material (+770%).

CONCLUSION
Cross-linking treatments greatly increased
UHMWPE wear resistance, especially when
aged. A reduction in tensile and toughness
properties was seen for all crosslinked materials
analyzed, although there were slightly better results for slow cool-rate heat treatments. There
was a slight improvement in toughness behavior
of materials heat-treated at 155 C. Tensile and
toughness tests, performed on selected materials before and after ageing, showed important

Address for correspondence:

Roberto Chiesa, PhD
Dip. Chimica, Materiali e Ingegneria Chimica "G. Natta"
Politecnico di Milano
Via Mancinelli, 7
20133 Milano - Italy
roberto.chiesa@polimi.it

27

Influence of heat treatment on structural, mechanical and wear properties of crosslinked UHMWPE

REFERENCES
1. Willert H, Bertram H, Buchhorn GH. Osteolysis in
alloarthroplasty of the hip The role of ultra-high
molecular weight polyethylene wear particles. Clin
Orthop 1990; 258: 95-107.
2. Amstutz HC, Campbell P, Kossovsky N, Clarke IC.
Mechanism and clinical significance of wear debrisinduced osteolysis. Clin Orthop 1992; 276: 7-18.
3. Cooper RA, McAllister CM, Borden LS, Bauer TW.
Polyethylene debris-induced osteolysis and loosening in uncemented total hip arthroplasty. J Arthroplasty 1992; 7: 285-90.
4. Encyclopedia of polymer science and technology, vol
6. Wiley & Sons 1976.
5. Muratoglu OK, Bragdon CR, OConnor DO et al.
Unified wear model for highly crosslinked ultra-high
molecular weight polyethylenes (UHMWPE). Biomaterials 1999; 20: 1463-70.
6. Oonishi H, Ishimaru H, Kato A. Effect of cross-linkage by gamma radiation in heavy doses to low wear
polyethylene in total hip prostheses. J Mater Sci
Mater Med 1996; 7: 753-63.
7. Chiesa R, Tanzi MC, Alfonsi S, Paracchini L,
Moscatelli M, Cigada A. Enhanced wear performance of highly crosslinked UHMWPE for artificial
joints. J Biomed Mater Res 2000; 50: 381-7.
8. Bragdon CR, OConnor DO, Muratoglu OK, et al. A
new polyethylene with undetectable wear at 12 million cycles. 24th Annual Meeting of the Society for
Biomaterials, 22-26 April, San Diego, California,
USA (1998), 2.
9. Shen FW, McKellop H, Salovey R. Improving the resistance to wear and oxidation of acetabular cups of
UHMWPE by gamma radiation crosslinking and
remelting. 24th Annual Meeting of the Society for
Biomaterials, 22-26 April, San Diego, California,
USA (1998), 3.
10. Wang A, Polineni VK, Essner A, Sun DC, Stark C,
Dumbleton JH. Effect of radiation dosage on the
wear of stabilized UHMWPE evaluated by hip and

28

11.

12.

13.

14.

15.
16.
17.
18.
19.

knee joint simulators. 23th Annual Meeting of the

Society for Biomaterials, 30 April-4 May, New Orleans, Louisiana, USA (1997), 394.
Jasty M, Bragdon CR, OConnor DO, Muratoglu OK,
Premnath V, Merrill EW. Marked improvement in
the wear resistance of a new form of UHMWPE in a
physiologic hip simulator. 23th Annual Meeting of
the Society for Biomaterials, 30 April-4 May, New Orleans, Louisiana, USA (1997), 157.
Laurent MP, Yao JQ, Bhambri SK, Gsell RA, Swarts
DF, Crowninshield RD. High cycle wear of highly
crosslinked UHMWPE acetabular liners evaluated in
a hip simulator. 46th Annual Meeting of the Orthopaedic Research Society, 12-15 March, Orlando,
Florida, USA (2000), 567.
Muratoglu OK, Bragdon CR, OConnor DO, et al.
Wear behavior and mechanical properties of highly
crosslinked UHMWPE preconditioned by 20 million
cycles of simulated gait. 6th World Biomaterials Congress, 15-20 May, Kamuela, Hawaii, USA (2000), 869.
Pascaud R, Evans T, McCullagh PJJ, FitzPatrick D.
Effects of batch to batch variations and test methodology on degree of cristallinity and melting temperature of UHMW-PE as measured by differential
scanning calorimetry. J Biomed Mater Res 1996; 32:
619-26.
Paul JP. Forces transmitted by joints in the human
body. Proc Inst Mech Eng (H) 1966; 181: 8-15.
Verdonschot N, Huiskes R. Dynamic creep behavior
of acrylic bone cement. J Biomed Mater Res 1995;
29: 575-81.
Bhateja SK, Duerst RW, Martens JA, Andrews EH. Radiation-induced enhancement of crystallinity in
polymers. J Macromol Sci-Rev M 1995; C35: 581-659.
Shinde A, Salovey R. Irradiation of ultra-high-molecular-weight-polyethylene. J Polym Sci: Pol Phys 1985;
23: 1681-9.
Shen FW, McKellop HA, Salovey R. Irradiation of
chemically crosslinked ultra high molecular
weight polyethylene. J Polym Sci: Pol Phys 1996;
34:1063-77.