J. Biomedical Science and Engineering, 2011, 4, 255-263 JBiSE
doi:10.4236/jbise.2011.44035 Published Online April 2011 (http://www.SciRP.org/journal/jbise/).
Published Online April 2011 in SciRes. http://www.scirp.org/journal/JBiSE
Uni- and multidirectional wear resistance of different
crosslinking degrees in UHMWPE for artificial joints
V. A. González-Mora1, M. Hoffmann1, R. Stroosnijder1, E. Espinar2, J. M. Llamas2,
M. Fernández-Fairén3, F. J. Gil3
1Institute for Health and Consumer Protection, Joint Research Centre, European Commission, Ispra, Italy;
2Grupo de investigación en Ortodoncia, Facultad de Odontología, Universidad d e Sevilla, Seville, Spain;
3CREB, Department Ciencia de Materiales e Ingeniería Metalúrgica, ETSEIB, Universidad Politécnica de Cataluña, Barcelona,
Email: francesc.xavier.gil@upc.edu
Received 20 December 2010; revised 17 February 2011; accepted 24 February 2011.
The aim of this work was to study the effect of
UHMWPE crosslinking on wear performance. Dif-
ferently treated UHMWPEs were studied by means
of unidirectional and multidirectional wear tests. Uni-
directional tests simulate total knee replacement and
multidirectional tests simulate total hip replacement
movements. The samples tested were observed by op-
tical and scanning electron microscopy in order to
determine wear mechanisms that explain the differ-
ent results obtained in the uni- and multidirectional
wear tests performed.
Keywords: Artificial Joints; Wear; Polymeric
When natural joints have to be replaced with artificial
materials, their tribological properties change as a result
of the inability of the materials to produce a permanent,
artificial lubricating film. Therefore, the materials used
for articulating components in an artificial joint are in-
evitably subject to wear. Furthermore, there is no ideal
bearing material that currently fulfils all the require-
ments of arthroplasty design [1,2]. It is therefore impor-
tant to ensure that wear is minimised to avoid possible
aseptic loosening following osteolysis due to particle-
initiated foreign body reaction [3-5].
The articulating surfaces in total joint replacement are
recognised as major sources of wear debris generation.
Other implant surfaces, specifically fixation surfaces,
may release additional wear debris during the in vivo
function. Thus, the origin of wear particles can be di-
vided into the prosthesis-bone interface and the prosthe-
sis-prosthesis interface, which can be designed to be
articulating or non-articulating. The amount of wear de-
bris from the former interface is low. This may be ac-
ceptable, but only if the debris does not mig rate to other
interfaces where it could contribute to third-body wear
The lack of an adequate standard hampers the com-
parison of studies carried out by va rious labo ratories and
the progress made in understanding the wear phenomena
that occurs in total join t replacements for classifying the
various UHMWPEs studied [7-10]. More must be done
than simply reviewing or improving the existing stan-
dards. Furthermore, there is a clear need to develop a
new standard for screening wear tests based on the latest
findings, such as multidirectional motion or lubricant
A pin-on-disk (POD) wear-test machine is a common
wear-test method that has been widely used to evaluate
the wear of polymers in biotribology. In a POD test, the
polymer, usually in the form of a pin, slides over the
surface of a rotating disk. Two basic configurations may
be employed: the pin is lo ad ed along th e main ax is of th e
disk either perpend icular o r parallel to its axis of rotation .
Hence, the contact area is either on the edge (horizontal
POD configuration) or on the face (vertical POD con-
figuration) of the disk. According to this definition, the
wear-test method proposed here was a horizontal POD
(hereinafter simply called POD I).
In the POD test device, a vertically positioned wheel
or ring rubs against a polyethylene pin below it. Figure
1(a) shows a diagram of the loading/motion configura-
tion. The contact geometry at the start of the test is
non-conformal, that is, in line contact. Therefore, these
wear-test devices would be more representative of a
knee-wear device than of a hip-wear device, for which
V.A. González-Mora et al. / J. Biomedical Science and Engineering 4 (2011) 255-263
Copyright © 2011 SciRes. JBiSE
Figure 1. Motion/loading con-
figuration of the POD wear test
machines. (a) Unidirectional. (b)
the contact geometry is conformal (also known as con-
There is no standard regarding this type of wear de-
vice nor are there any similar studies in the literature.
Thus, a comparison and analysis of the test conditions
proposed cannot be made. Due to its configuration, the
closest available standard would be the ASTM G137-97
standard, which has identical contact geometry and is
also intended for classifying the resistance of plastic
materials in sliding wear. However, this standard does
not cover biotribological conditions.
The multidirectional POD (POD II) wear-test method
is based on the former unidirectional POD (POD I)
wear-test method. However, in the POD II test method,
the pin also rotates (see Figure 1(b)) and, as conse-
quence of the rotation of the pin, this device displays a
biaxial (i.e. multidirectional) motion. In the literature, no
wear device similar to this can be found, which makes it
a unique screening wear-test device. In theory, the tasks
undertaken using this wear-test method are the same as
those undertaken using the former POD I. Additionally,
the effect of the motion type in the wear resistance of the
UHMWPE may be studied by means of the POD II tests.
The test using the POD test method was performed as
follows. The disk rotated continuously against a UHM-
WPE pin placed beneath it. A load of ~150 N (15 kg)
was applied to the pins, which gave a maximum Hert-
zian contact pressure (p0) of ~5 MPa. The relative sur-
face velocity was 100 rpm for the disk; in the case of the
multidirectional test, the pin velocity was 99 rpm. The
disk had a frequency of 1 Hz. The wear of the UHM-
WPE pins was mainly determined by profilometric mea-
surements. Weight loss measurements were also perfo rmed
after the completion of the test. The test length was 34
Distilled water was added to the test lubricant during
the test to compensate for water evaporation. A solution
consisting of bovine serum and distilled water was used
as the test lubricant, which had a total protein concentra-
tion of 30 mg/ml. The serum was purchased from
Sigma-Aldrich (calf serum, bovine donor; product no.
C9676). The fluid adsorption rate of the UHMWPE pins
was determined using an additional control pin, which
was loaded in exactly the same way as the UHMWPE
pins in the RPOF machine, but no motion was applied.
The cleaning and dr ying of the UHMWPE pins was per-
formed according to the ASTM 1715 standard. Weighing
was carried out using a Mettler Toledo AT261-Delta-
Range® microbalance with an accuracy of 10 µg.
The disks were made of forged CoCrMo alloy (pur-
chased from Firth Rixson Superalloys Ltd., Derbyshire,
England). The disks were 88 mm in diameter and 10 mm
thick. The roughness of the disks was determined by
laser profilometer, which yielded a value of Ra = 0.01
µm. The pins were manufactured from a bar of UHM-
WPE GUR1050, which was 13 mm in length and 9 mm
in diameter. Four different treatments on the UHMWPE
material were studied.
Crosslinked I (-sterilised + stabilised I) with the
following treatment:
By irradiation with 100 kGy of gamma ray in
air (sterilisation)
By the McKellop heat treatment at 155˚C for
V.A. González-Mora et al. / J. Biomedical Science and Engineering 4 (2011) 255-263
Copyright © 2011 SciRes. JBiSE
72 hours in nitrogen (stabilisation)
Crosslinked II (-sterilised + stabilised II) with the
following treatment:
By irradiation with 100 kGy of gamma ray in
air (sterilisation)
By heat treatment under water at 130˚C in H2O
for 72 hours (stabilisation)
Sterilised with a standard 25 kGy (2.5 Mrad) of
gamma radiation in air
For each UHMWPE material, three samples were
tested. A total of 12 wear tests were performed. Test con-
ditions and materials are summarised in Table 1.
The wear results, given in volumetric wear units (mm3),
are calculated from the average weight loss (mg) re-
corded for three specimens of each UHMWPE material.
A higher weight loss of the UHMWPE material repre-
sents greater wear in the sample. A graph of the wear
results is shown in Figure 2.
3.1. Unidirectional Wear Test
For the unidirectional test, the results showed higher
wear for the irradiated and crosslinked UHMWPEs
(XLPEs) than for the unirradiated UHMWPE material.
The UHMWPE wear obtained for the irradiated materi-
als and XLPEs were in the same order, yielding a 1.5 to
1.8 fold increase with respect to the unirradiated UHM-
WPE. A high reproducibility of the wear tests was achi-
eved, in which the low standard deviation in the weight
loss of the specimens did not ever exceed 7%. The re-
sults show that the difference between crosslinked mate-
rials is not statistically significant (p < 0.05), and that the
two crosslinking treatments seem to have a similar effect
on the wear resistance of the UHMWPE. The results also
demonstrate that the unidirectional motion applied is
unable to differentiate between irradiated
Table 1. Test conditions of the POD wear tests.
Test parameter Value
Contact geometry Cylinder-on-flat (non-conformal)
Frequency 1 Hz
Relative surface velocity 100 rpm
Contact area Line
Load applied ~150 N (15 kg)
Contact stresses 5 MPa
Test length 34 Km (123 000 d i s k rotations)
Lubricant 30 mg/ml initial protein content
Temperature Room
Counterface component CoCrMo alloy
UHMWPE component
(GUR 1050)
Crosslinked I (-sterilised + stabilised I)
Crosslinked II (-sterilised + stabilised II)
Figure 2. Average wear of the UHMWPE pins in the unidirec-
tional POD wear tests.
materials (irradiated UHMWPEs and XLPEs), even if
they are quite different. Due to the four-fold irradiation
dose (100 kGy), XLPEs have a higher degree of cross-
linking compared with UHMWPEs that are irradiated
with 25 kGy. Despite this difference in the degree of
crosslinking, when a unidirectional sliding motion is
applied they display a similar degree of resistance. The
significance of the results obtained by employing unidi-
rectional motion is questionable, because wear tests us-
ing this kind of motion only seem to be able to detect
differences in the wear behaviour of unirradiated and
irradiated UHMWPEs, while differences between dif-
ferent degrees of crosslinking cannot be detected.
Questions may arise regarding the influence of load
on UHMWPE wear, for example, if a difference in the
wear of XLPEs were encountered whenever higher loads
were applied. In ou r opinion, the use of hig h e r loa ds may
result in higher UHMWPE wear for all the materials
studied here, but the classification of materials would
remain unchanged. Studies [11,12] focusing on the effect
of load on UHMWPE wear have concluded that this is
not a key parameter when an attempt is made to classify
The surface of the pins was observed by means of op-
tical microscopy (OM), as can be seen in Figure 3(a). In
Figure 3(b), the surface resulting from the machining
process shows the marks left behind by the machining
tool’s cutting action. The main features are the scratches
caused by the action of the disk sliding over the pin’s
surface, which leaves a typical unidirectional lay struc-
ture, in line with the sliding direction of the disk on the
pin’s surface (see Figure 3(c)). To the naked eye, the
wear zone appears to be polished. Furthermore, besides
the scratches mentioned above, the continuou s sliding of
the disk on the pin’s surface in the wear zone formed a
ripple-like microstructure or lay of the fibres that make
up the UHMWPE. The ripples are perpendicular to the
V.A. González-Mora et al. / J. Biomedical Science and Engineering 4 (2011) 255-263
Copyright © 2011 SciRes. JBiSE
V.A. González-Mora et al. / J. Biomedical Science and Engineering 4 (2011) 255-263
Copyright © 2011 SciRes. JBiSE
Figure 3. (a) Image of the unirradiated UHMWPE pin after the unidirectional wear test. The black arrow shows the direction of
sliding. (b) Detail of the machined surface. (c) Detail of the worn surface of unirradiated UHMWPE. Note also the scratches par-
allel to the sliding direction. (d) Detail of the worn surface of irradiated UHMWPE. (e) Detail of the worn surface of XLPE I. (f)
Detail of the worn surface of XLPE II. (g) to (j) Images obtained by SEM.
sliding direction. It could be concluded that the micro-
structure of the UHMWPE fibres clearly falls into align-
ment with the sliding direction.
Optical micrographs of the worn surfaces revealed the
ripple-like structure on all the p ins studied (Figure 3(d)).
However, the size of the ripples was different if the pin
was made of an XLPE material (Figures 3(e) and 3(f)).
The ripple-like structure is clearly smaller compared
with the microstructure found in th e unirradiated or irra-
diated UHMWPEs. The ripples on XLPEs are smaller
because their fibres are smaller in comparison to non-
crosslinked UHMWPEs. This is a consequence of the
heat treatment that XLPEs undergo after irradiationper-
formed at 155˚C in the case of these materials-which
acts as a remelting process for the UHMWPE fibres.
Another feature of the worn surfaces is that the irradi-
ated UHMWPEs and XLPEs have a greater number of
scars than the unirradiated UHMWPEs, and that the
scars are shallower. Both the morphology of the ripples
and the scratches displayed a lower degree of deforma-
tion in the irradiated UHMWPEs and XLPEs, due to the
irradiation-induced crosslinking process.
The observation by means of SEM focused on the
formation of UHMWPE particles that detach from the
pin surface, which gives rise to wear debris. The scan-
ning electron micrographs of the unirradiated (Figure
3(g)), irradiated (Figure 3(h)), XLPE I (Figure 3(i)) and
XLPE II (Figure 3(j)) materials are shown below.
The SEM observations (Figures 3(g) to 3(j)) show
that the UHMWPE pins exhibit a cracked surface texture.
This texture shows up as micro-cracks, which are pre-
sent in every direction but most frequently occur be-
tween ripples, thus highlighting the ripple-like micro-
structure of the UHMWPE when it is observed under an
optical microscope. These micro-cracks are, however, a
result of gold sputtering, which is necessary to bring
about conductivity on the UHMWPE surface. They do
not show up as micro-cracks on the UHMWPE surface.
This effect could not be avoided even if the sputtering
periods were very short.
Besides the ripple-like microstructure, other features
can be observed, such as the particle formation in the
form of fibrils on the worn surfaces. The study of the
particle formation is essential to understanding the wear
processes that occur, as a hypothesis has been advanced
that wear particles may be liberated from the articulating
surface after the cyclic accumulation of a critical amount
of plastic strain.
There is a greater degree of fibril formation in the
unirradiated material than in the irradiated material, but
less so in the case of XLPEs. For non-XLPEs, fibrils are
placed parallel to the sliding direction and may extend
over several ripples. In the case of XLPEs, rounded par-
ticles smaller than those that occur in non-XLPEs are
formed rather than fibrils. The particle formation in the
XLPEs once again corroborates the fact that there tends
to be less deformation in the XLPEs in comp arison with
irradiated and unirradiated UHMWPEs. In Figure 3(j),
the concentration of plastic strain on the polyethylene
surface in the XLPEs can be observed.
With regard to the results for weight loss, the particle
formation in the XLPEs should have been higher than in
the non-XLPEs, since UHMWPE wear was also higher.
At this point, it is worth remembering the phenomenon
of irrecoverable, permanent strain resulting from me-
chanical loading, which is termed plasticity and holds
clues as to the wear and mechanical loading history of
UHMWPEs. The higher pr oduction of wear debr is in the
XLPEs may be explained by the fact that the particles
that form on the XLPEs immediately become detached.
This is due to the fact that they are able to bear a lower
strain concentration (5.6) than non-XLPEs particles,
V.A. González-Mora et al. / J. Biomedical Science and Engineering 4 (2011) 255-263
Copyright © 2011 SciRes. JBiSE
which coincides with the appearance of rounded and
smaller XLPE particles. However, non-XLPEs have a
higher deformation capacity (or plasticity), allowing par-
ticles to retain a higher accumulation of plastic strains
and resulting in less particle detachment and, conse-
quently, less weight lo ss. This coincides with the known
fact that the molecular orientation of UHMWPE occurs
under unidirectional sliding conditions, including ani-
sotropy in the UHMWPE and orientation hardening in
the slidi ng di rect ion [11]. In concl usi on, UHM WPE s with a
greater capacity to deform locally (i.e. with greater plas-
ticity) present higher wear resistance and less weight
loss under unidirectional conditions. For the same rea-
sons, there is less wear in unirradiated UHMWPE than
in irradiated UHMWPE, as this and other studies have
shown, since the latter possesses lower plasticity due to
sterilisation (irradiation process).
3.2. Multidir ecti onal Wear Test
The wear results of the UHMWPE samples (pins) ob-
tained using the POD II (multidirectional) test method
are shown in Figure 4.
The wear results show a higher wear rate for the irra-
diated and unirradiated UHMWPE materials compared
to the XLPEs. That is, the un irradiated UHMWPE mate-
rials and the irradiated UHMWPE materials in particular
perform worse than the crosslinked materials. In fact, the
wear observed in the irradiated and unirradiated UH-
MWPEs was 3.5 and 5 times higher, respectively, than
that in XLPE II. Moreover, wear in the irradiated mate-
rial was 1.43-fold higher than that in the unirradiated
UHMWPE. XLPE I and II displayed similar wear resis-
tance, although the latter showed slightly lower wear
than XPLE I, which is statistically significant (p < 0.01).
A high reproducibility of the wear tests was ach ieved for
all the materials tested, which is expressed by the low
standard deviations (SD < 5%).
Figure 4. Average wear of the UHMWPE pins in the multidi-
rectional POD wear tests.
Contrary to the POD I test method, in which multidi-
rectional sliding conditions were employed, the POD II
test method showed that multidirectional motion is able
to differentiate between irradiated materials in terms of
their wear rates, i.e. between XLPEs and irradiated
UHMWPEs. The results also demonstrated that under
multidirectional sliding conditions XLPE materials im-
prove the tribological resistance of UHMWPEs, which
exhibit 4.4. to 5 times less wear compared with the
standard irradiated UHMWPE component. For example,
assuming that an acetabular cup made of standard irradi-
ated UHMWPE has a common mean lifetime of 10 to 15
years, the higher wear resistance of XLPEs could in-
crease the mean lifetime by 45 to 66 years.
This comparison between uni- and multidirectional
results leads to two conclusions:
a) Multidirectional sliding motions cause greater UHM-
WPE wear than unidirectional sliding motions.
b) The classification of the materials tested shows an
inverse trend under unidirectional and multidirectional
sliding motions.
The two classes of wear results-the inverse classifica-
tion and the order of magnitude of UHMWPE wear-
coincide with the results obtained in stud ies found in the
literature. This supports the POD wear-test method ap-
plied in this work as a suitable screening wear test for
the evaluation of UHMWPE wear.
The multidirectional wear results performed on the
POD machine also show that the XLPEs are an optimal
material for acetabular bearing components in total hip
replacements, since UHMWPE wear can be significantly
reduced with respect to the standard irradiated (25 kGy)
UHMWPE. The results under unidirectio nal sliding con-
ditions show, however, that the XLPEs are subject to
more wear compared with the unirradiated UHMWPE.
Therefore, in situations in which the unidirectional slid-
ing motion is the main type of motion between the ar-
ticulating components, as encountered in most current
total knee replacement designs, XLPEs should not be
used. It is now recognised that the different kinematics
in the hip and knee can lead to a different wear mecha-
nism surface and wear rate of the UHMWPE component
[12]. For total hip replacements, it is now widely ac-
cepted that wear is related to the mechanical response of
UHMWPE materials under multidirectional conditions.
In total hip replacements, variation in the direction of the
velocity vector leads to cross shearing on the strain-
hardened polyethylene and accelerates wear. However,
for total knee replacements, the major contribution made
to sliding motion seems to be unidirectional motion,
depending on the prosthesis design, and thus the re-
sponse of UHMWPE under unidirectional conditions
seems to be the most significant. Retrieved total knee
components present scratches on the UHMWPE com-
V.A. González-Mora et al. / J. Biomedical Science and Engineering 4 (2011) 255-263
Copyright © 2011 SciRes. JBiSE
ponent that are predominantly parallel to the wear slid-
ing direction [6,7]. Thu s, in total knee replacements with
a unidirectional motion, the wear may be lower due to
the orientation and strain hardening of the UHMWPE in
the direction of motion. Two factors affecting UHMWPE
wear and behaviour that have not yet been satisfactorily
addressed must be taken into account: the high variabi-
lity in knee designs and the much higher stress contacts
in comparison with hip de signs.
The optical microscopies at low magnification are
presented in Figures 5(a) to 5(d). They show that the
unirradiated and irradiated UHMWPEs exhibit greater
wear damage when they are compared to XLPEs, which
are consistent with the weight loss results. It should be
Figure 5. (a) Image of the surface of the unirradiated UHMWPE pin after the multidirectional wear test, (b) for irradiated
UHMWPE, (c) for XLPE I, (d) XLPE II, (e) to (h) Details of the microstructure observed by optical microscope. (i) to (l) Details
of the microstructure observed by SEM.
V.A. González-Mora et al. / J. Biomedical Science and Engineering 4 (2011) 255-263
Copyright © 2011 SciRes. JBiSE
Figure 6. Region of the unirradiated UHMWPE pin: SEM micrographs after multidirectional wear test.
noted that parallel scratches found on the pin surfaces in
the unidirectional tests (POD I) are not present here and
only multidirectional shallow scratches can be seen.
Optical microscopy images at higher magnification
were taken at the midpoint of the pin’s diameter (see
Figures 5(e) to 5(h)). The central plateau was not con-
sidered because the optical microscope was unable to
observe any features other than the flatness of this area.
The figures reveal a similar ripple-like microstructure
for all the pins studied. In this case, the ripples are less
aligned than those found in the unidirectional wear tests.
Again, the microstructure of the ripples is more evident
in the unirradiated and irradiated UHMWPEs than in the
XLPEs, which reinforces the theory that the XLPEs have
lower plasticity, as discussed in the results on unidirec-
tional wear. The ripples are perpendicular to the sliding
direction, which under the multidirectional sliding con-
dition yielded results in which the ripples were radially
oriented, as if the sliding directions were from the centre
of the pin to its edge.
The SEM observation was focused on the formation
of UHMWPE particles that detach from the pin surface
to produce wear debris. The scanning electron micro-
graphs of the unirradiated (Figure 5(i)), irradiated (Fig-
ure 5(j)) XPLE I (Figure 5(k)) and XLPE II (Figure
5(l)) materials are shown below.
The ripple-like microstructure identified by optical
microscopy is present in the area outside the central pla-
teau, both for the unirradiated and irradiated UHMWPEs.
The central plateau of the pin shows an extremely ho-
mogeneous and flat microstructure with no visible rip-
ples. However, there is particle and fibril formation on
the central plateau. The formation of fibrils was also
observed on the rest of the worn surface. As can be ob-
served in Figure 6, all the fibrils are oriented along the
sliding direction and are perpendicular to the ripple-like
microstructure (radially from the pin’s centre to the pin’s
edge), and these fibrils can extend over several ripples.
Fibril formation is more marked for the unirradiated th an
for the irradiated material, followed by the crosslinked
For the XLPEs, the homogeneous ripple-like micro-
structure can be seen all over the pin’s surface. Fibril
formation is less marked for the XLPEs than for the
unirradiated and irradiated UHMWPEs. Compared with
the non-crosslinked UHMWPEs, XLPEs show much
less particle formation and the fibrils are smaller. As far
as the unidirectional tests are concerned, the fibrils are
much rounder and smaller, which under the SEM appear
as white particles forming on the ripples of the micro-
structure. As explained for the unidirectional wear tests,
this particle formation in XLPEs is caused by th eir lower
Based on the size of the fibrils forming on both non-
crosslinked UHMWPEs and XLPEs, the size of the par-
ticles forming the UHMWPE debris can be estimated.
Figures 5(i) to 5(l) and Figure 6 show that the size of
the UHMWPE particles that detach from the worn sur-
faces ranges from a submicron to more than one micron.
It has been found that in total hip replacements most
UHMWPE particles are less than one micron in length
[13]. The findings of studies of wear particles retrieved
from periprosthetic tissues and analyses of worn poly-
ethylene surfaces are consistent with an average particle
size in the 0.5 micrometer diameter range [3,7].
[1] Miller, D.A., Herrington, S.M., Higgins, J.C. and Schroe-
der, D.W. (1995) UHMWPE polyethylene in total joint
replacement: History and current technology. In: Wise,
D.L., et al. Eds., Encyclopaedic Handbook of Biomate-
rials and Bioengineering, Part B: Applications, Marcel
Dekker, Inc., New York, 665-688.
[2] Greer, K.W., Hamilton, J.V. and Cheal, E.J. (1995) Poly-
ethylene wear in orthopaedics. In: Wise, D.L., et al. Eds.,
Encyclopaedic Handbook of Biomaterials and Bioengi-
neering, Part B: Applications, Marcel Dekker, Inc., New
York, 613-638.
V.A. González-Mora et al. / J. Biomedical Science and Engineering 4 (2011) 255-263
Copyright © 2011 SciRes. JBiSE
[3] Murray, D. and Rushton, N. (1990) Macrophages stimu-
late bone resorption when they phagocytose particles.
Journal of Bone and Joint Surgery, 72-B, 988-992.
[4] Howie, D. and McGee, M. (1996) Wear and osteolysis in
relation to prostheses design and materials. Medical Ap-
plications of Titanium and Its Alloys (ASTM STP 1272).
[5] McGee, M., Howie, D., Neale, S., Haynes, D. and Pearcy,
M. (1997) The role of polyethylene wear in joint failure.
Proceedings of the Institution of Mechanical Engineers—
Part H, 211, 65-72. doi:10.1243/0954411971534692
[6] Green, T., Fisher, J., Stone, M., Wroblewski, B. and Ing-
ham, E. (1998) Polyethylene particles of a “critical size”
are necessary for the induction of cytokines by macro-
phages in vitro, Biomaterials, 19, 2297-2302.
[7] Greer, K.W., Hamilton, J.V. and Cheal, E.J. (1995) Poly-
ethylene wear in orthopaedics. In: Wise, D.L., et al. Eds.,
Encyclopaedic Handbook of Biomaterials and Bioengi-
neering, Part B: Applications, Marcel Dekker, Inc., New
York, 613-638.
[8] Plitz, W. and Walter, A. (1994) Tribological aspects of
metal/polymer couplings. In: Buchhorn, G. and Willert,
H.G., Eds., Technical Principles, Design and Safety of
Joint Implants, Hogrefe & Huber Publishers, Cambridge,
[9] Algan, S.M. and Horowitz, S.M. (1995) Biology of asep-
tic loosening of the cemented arthroplasty. In: Wise, D.L.,
et al. Eds., Encyclopaedic Handbook of Biomaterials and
Bioengineering, Part B: Applications, Marcel Dekker,
Inc., New York, 773-797.
[10] Viceconti, M., Cavallotti, G., Andrisano, A. and Toni, A.
(1996) Discussion on the design of a hip joint simulator.
Medical Engineering & Physics, 18, 234-240.
[11] Ben Abdallah, A. and Treheux, D. (1991) Friction and
wear of utrahigh molecular weight polyethylene against
various new ceramics. Wear, 142, 43-56.
[12] Suh, N., Mosleh, M. and Arinez, J. (1998) Tribology of
polyethylene homocomposites. Wear, 214, 231-236.
[13] Streicher, M. (1995) Tribology of artificial joints. In:
Morscher , E., Ed., Endoprosthetics, Springer Verlag, Berlin.