Mechanical performance of electron-beam-irradiated UHMWPE in vacuum and in air

Mechanical Performance of Electron-Beam-IrradiatedUHMWPE in Vacuum and in Air

A. M. Visco,1 L. Torrisi,2 N. Campo,1 U. Emanuele,2 A. Trifiro,2 M. Trimarchi2

1 Industrial Chemistry and Materials Engineering Department, Engineering Faculty, University of Messina,Messina, Italy

2 Physics Department, Science Faculty, University of Messina, Messina, Italy

Received 2 March 2007; revised 3 April 2008; accepted 26 May 2008Published online 5 September 2008 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jbm.b.31187

Abstract: Ultrahigh molecular weight polyethylene (UHMWPE) was modified by a 5-MeVenergy electron beam at different temperatures before, during, and after irradiation, both inair and in high vacuum. Wear resistance, hardness, and tensile strength of irradiatedpolyethylene were compared with those of untreated one. Physical analyses (like infraredspectroscopy and calorimetric analysis) were carried out to investigate about the changes inthe material induced by irradiation. Experimental results suggested that structural changes(double bonds, crosslinks, and oxidized species formation) occur in the polymer depending onthe environmental conditions of the irradiation. Mechanical behavior is related to thestructural modifications. A temperature of 1108C before, during, and after the in vacuumirradiation of UHMWPE produces a high amount of crosslinks and improves polymeric tensileand wear resistance, compared to that of the untreated material. ' 2008 Wiley Periodicals, Inc.

J Biomed Mater Res Part B: Appl Biomater 89B: 55–64, 2009

Keywords: polyethylene (UHMWPE); radiation; oxidation; crystallinity; mechanicalproperties

INTRODUCTION

UHMWPE (ultrahigh molecular weight polyethylene) rep-

resents a polymer with special properties in terms of me-

chanical resistance, biocompatibility, chemical inertia, and

ductility. It has been used for over 30 years in different

fields, such as microelectronics, chemistry, engineering,

biology, and medicine.1

Several studies have been concerned with UHMWPE

improving the properties of UHMWPE by using irradiation

processes.2–6 Irradiation effects induced on UHMWPE are

difficult to predict and very complex, despite the simple

polymeric structure.

Generally, free radicals are produced during the polyeth-

ylene irradiation; they create new chemical bonds as well

as chain scissions.7

In an inert environment or under vacuum, as well as the

presence of air, free radicals can react with one another to

form intermolecular and/or intramolecular carbon–carbon

bonds. These events prevail over chain scissions in polyeth-

ylene irradiated in vacuum or in an inert environment. If

chains mobility is enough, intermolecular crosslinks prefer-

entially occurs. Crosslinks formation enhances the material

wear resistance since a polymeric network better opposes

to the wear stress.8 When irradiation is performed in the

presence of air, free radical easily react with oxygen result-

ing in broken chains that contain oxidized species. Chain

scission reactions decrease the material performance when

compared with untreated UHMWPE.9

So, irradiation does not always improve polymer quality;

sometimes irradiation worsen it. For example, the gamma

radiation sterilization in air at a dose of 25 kGy is com-

monly performed on medical prostheses. This process

improves the wear resistance of UHWMPE immediately af-

ter irradiation. However, gamma radiation sterilization also

enhances the delamination wear after long-term implanta-

tion, as a consequence of the slow oxidation process initi-

ated by long-lived free radicals in the presence of

oxygen.10

When UHMWPE is exposed to irradiation in an inert

environment and then aged in air, chain scission prevails

over crosslinking formation.11 Also in this case, oxygen is

able to enter readily into the material and react with the

long-lived radicals that have not had the opportunity to

crosslink.12

These studies clearly suggest the necessity to perform

the irradiations without the presence of oxygen before,

during, and after modification with irradiation. It is also

Correspondence to: A.M. Visco (e-mail: [email protected])

' 2008 Wiley Periodicals, Inc.

55

necessary to stabilize the free radicals by heating the poly-

mer just after irradiation, in order to allow free radicals

reaction and create crosslinks instead of chain scission. Pol-

ymeric structure should therefore be stabilized in this way

so that the subsequent ageing in air does not affect its

structure.13

Ion irradiations in vacuum have been employed with

success to induce crosslinks in the polymeric surface,

increasing its hardness and wear resistance.14 However,

energetic electrons have ranges in the order of centimeters

and may introduce energy to the polymer bulk. The dose

and dose rate should be controlled accurately in order to

find the best conditions for polymer modification without

worsening its chemical and physical properties.15

In this work, UHMWPE was electron beam irradiated

both in air and under high vacuum with a dose of 100 kGy

in order to avoid its embrittlement.15

The nonirradiated UHMWPE was mechanically com-

pared with a polymer heated at different temperatures

before, during, and after electron irradiations. The aim was

to study the best irradiation conditions to enhance

UHMWPE wear resistance without affecting its strength

and ductility. Some physical investigations, like infrared

spectroscopy and calorimetric analysis, were also per-

formed in order to study the structural changes induced by

irradiation and to connect these changes to mechanical

performance.

MATERIALS AND METHODS

Samples Preparation

A Ticona UHMWPE resin GUR 1020 (average molecular

weight of 2–4 3 106 g/mol, density of 0.93 g/cm3, without

calcium stearate) was employed to obtain 100 mm 3100 mm and 1-mm thick sheets. UHMWPE samples were

prepared by compression molding in a laboratory press.

The polymer powder was kept at 2008C for about 15 min

at 20 MPa pressure according to the conditions used by

Suarez and de Biasi.13 Samples were prepared with two

geometries. The first specimen geometry employed for

wear testing and IR spectroscopy was a 20 mm 3 20 mm,

1 or 0.2 mm thickness. The second specimen geometry

used for tensile stresses was made according to the ASTM

638 M-3 international protocols (60 mm total length,

10 mm useful length, 2.5 mm minimal width, 1 mm thick-

ness). This second specimen geometry was obtained by

using a manual DGT System sample cutting press.

Samples Irradiation and Stabilization

A1 kW autofocusing standing wave electron LINAC (linear

particle accelerator) from the Physics Department, Messina

University, Italy was employed to accelerate electrons at 5

MeV energy with the following parameters: a repetition

rate of 1–300 Hz, a pulse duration of 3 ls, a peak power

of 1 MW, a radio frequency of 3 GHz, a max peak current

of 200 mA, and a beam aperture of 12 mm. All the irradia-

tion treatments of polyethylene were performed at 60 mA

current with a 3 Hz repetition rate. The average stopping

power of 5 MeV electrons in UHMWPE was 185 keV/mm

and the range was 27 mm. An accurate measurement of the

radiation dose per unit electron current was performed, at

different distances from the electron beam exit window, by

means of gafchromic films. This allowed us to measure the

total dose absorbed by a given sample depending on the

total electron charge. The sample was then collected by

charge integrator, coupled with a toroidal ferrite, which

continuously monitored the beam electron current stability.

A polymeric slab of 1-mm thickness was placed in a

suitable holder in a small vacuum chamber (Figure 1) for

in vacuum irradiations. The base pressure without irradia-

tion was 5 3 1027 mbar, whereas during irradiation it was

about 5 3 1026 mbar. The beam entered the vacuum

chambers going through a 50-lm-thick titanium window,

which was thick and rigid enough to maintain a vacuum

without measurably affecting electron beam energy. A

proper electric heating slab, ranging from room temperature

to 2008C, was coupled with the sample. The sample tem-

perature was measured by a thermocouple and continuously

monitored during the irradiation in order to keep a constant

temperature. The sample irradiations were performed at 50,

100, and 150 kGy, with a dose rate of 275 Gy/s both in air

and in vacuum.

UHMWPE was heated at different temperatures before

(Tb), during (Ti), and after (Ta) electron irradiation, within

a 25–1508C range. Every sample was codified with a letter

and a number. The letter refers to the treatment type per-

formed on UHMWPE: UT stands for untreated, A stands

for in air irradiated, and V stands for in vacuum irradiated.

The number refers to the Tb, Ti, and Ta values, given in

detail in Table I.

A1, A2, and A3 differed from V1, V2, and V3 for what

concerns the irradiation environment (air or vacuum). A1

Figure 1. Schematic of the vacuum chamber and electron beam

irradiation system.

56 VISCO ET AL.

Journal of Biomedical Materials Research Part B: Applied Biomaterials

and V1 differed from the other samples for what concerns

the temperature before and during the irradiation, that was

of 258C.A3 and V3 differed from A1, A2, V1, and V2 samples

for what concerns the annealing absence (performed after

the irradiation at Ta 5 1108C/60 min).

Finally, V4 and V5 differed from the V1, V2, and V3

samples for what concerns the temperatures before during

and after irradiation in vacuum, that were higher (130 or

1508C). The cooling rate to room temperature was of

0.78C/min for all the samples.

Characterization Analyses

Mechanical and physical test analyses were performed on

both the untreated and the electron-irradiated polymer.

Tensile stresses were applied to the UHMWPE samples

(60 mm 3 2.5 mm and 1 mm thick) at 258C by a universal

testing machine (LLOYD LR 10K, Engineering Faculty,

Messina University, Italy) with a crosshead speed of 1 mm/

min. For each irradiation dose 10 specimens were tested in

order to give the average value. The measurements gave

values resulting in the following parameters: the tensile

module, Et (MPa); the tensile yield strength, ry (MPa); the

ultimate tensile strength, ru (MPa); and the elongation at

break, eu (%). The work to fracture is a parameter that

gives an estimate of the material toughness. It was obtained

by the integration of the area under the stress–strain curves.

The first two parameters were correlated by the

equation:

Et ¼ lime!0

rye

ð1Þ

where e is the relative elastic excursion.

Wear measurements were performed by a ‘‘pin on disc’’

test machine (Engineering Faculty, Messina University,

Italy) to evaluate the weight loss due to friction material

(20 mm 3 20 mm and 1 mm thick). The samples were

placed on a rotating plate and stainless steel AISI 316 L

pin was pressed against the polymer surface generating a

pressure of 10 MPa, without any lubrication liquid. The

standard radius of the pin on disc test machine was

4.25 mm. The simulations were carried out at the speed of

120 rotations per minute (rpm). The wear rate (Wr) was

determined by dividing the weight loss (estimated from the

weight difference before and after the wear stress, by using

a microbalance) by the number of cycles (expressed in mil-

lion cycles) based on the following formula:

Wr ¼ Pi � Pf

N cycles=106ð2Þ

where Pi is the initial weight (g) and Pf is the final weight

after the wear stress due to the number of cycles (Ncycles). Each sample was polished with ethanol before and

after the wear test.

FTIR spectroscopy was performed on UHMWPE film

(20 mm 3 20 mm and 0.2 mm thick) by using a spectrom-

eter (Thermo Nicolet ‘‘Evolution 500’’ instrument, Engi-

neering Faculty, Messina University, Italy). The presence

of four chemical bonds was investigated. They were as

follows: vinyl group CH2¼¼CH�� (at wavelength of

908 cm21), trans vinylene C¼¼C (at wavelength of 965

cm21), carboxylic C¼¼O (at wavelength of 1716 cm21), and

hydroperoxides ��O��OH (at wavelength of 3550 cm21).

It is possible to quantify the reaction types originating

from the different treatments through the determination of

reaction indexes (RIvinyl, RITransvinylene) and oxidation

indexes (OICarbonyl,16 OIHydroperoxides) calculated by the fol-

lowing formulas:

RIvinyl ¼ Avinyl

A1370

ð3Þ

RITransvinylene ¼ ATransvinylene

A1370

ð4Þ

OICarbonyl ¼ ACarbonyl

A1370

ð5Þ

OIHydroperoxides ¼ AHydroperoxides

A1370

ð6Þ

where Avinyl is the CH2¼¼CH�� peak area, within the 895–

925 cm21 range, ATransvinylene is the C¼¼C peak area, within

the 950–980 cm21 range, ACarbonyl is the C¼¼O group peak

area, within the 1690–1750 cm21 range, AHydroperoxides is

the ��O��OH group peak area, within the 3520–

3580 cm21 range, A1370 is the polymer crystalline region

peak area, within the 1340–1390 cm21 range.

Microhardness measurements were performed by apply-

ing a load of 0.49 N for 10 s, on samples with 2.5 mm 310 mm surface dimension (Shimatzu DUH-200 instrument,

Engineering Faculty, Messina University, Italy).

Changes in crystallinity content and melting temperature

were assessed by heating samples (n 5 3) in a differential

TABLE I. Condition of Temperature Performed Before,During, and After UHMWPE Irradiation with 100 kGyElectron Beam in Air and in Vacuum

Sample

Code

Irradiation Conditions

Before During After

Tb(8C)

Time

(min) Ti (8C)Ta(8C)

Time

(min)

UT – – – – –

A1 25 0 25 110 60

A2 110 0 110 110 60

A3 110 0 110 – –

V1 25 0 25 110 60

V2 110 0 110 110 60

V3 110 0 110 – –

V4 130 0 130 130 60

V5 150 0 150 150 60

57MECHANICAL PERFORMANCE OF ELECTRON-BEAM-IRRADIATED UHMWPE


scanning calorimeter (DSC mod.Q-100 supplied by TA

Instruments, Engineering Faculty, Messina University,

Italy). Specimens were weighed with a microbalance and

placed in aluminum sample pans. The sample and the refer-

ence were then heated from 20 to 2208C at a heating rate

of 108C/min. Samples crystallinity was determined by inte-

grating the enthalpy peak from 20 to 2208C and normaliz-

ing it with the enthalpy of melting of 100% crystalline

polyethylene, 291 J/g.2 Lamellar thickness was calculated

according to the Thomson-Gibbs equation.17

RESULTS

All the samples studied were mechanically compared by

means of tensile tests whose average parameters, with

standard deviations, are given in detail in Table II.

Mechanical comparisons can be made between the UT,

A1, A2, A3, V1,V2, V3 samples whose stress/strain curves

are shown in Figure 2(a,b). The tensile module of all the

samples increased while elongation at break decreased

compared to those of the untreated UHMWPE. Similarly,

work to fracture generally decreased with respect that of

UT, although V2 showed the greatest work to fracture

value among all the samples. Besides, work to fracture of

V-series samples was generally greater than those of A-se-

ries ones. Yielding tensile strength and ultimate tensile

strength of irradiated sample were influenced by irradiation

environment. They both decreased in the presence of air

and increased in vacuum with respect to that of UT sample,

although A2 and V3 did not follow this trend. It must be

highlighted that V2 showed the highest ry and ru values

among all the samples. These results suggested that:

1. irradiation treatment generally stiffened the UHMWPE;

2. in vacuum irradiated samples showed a better mechani-

cal performance compared to that of air irradiated sam-

ples;

3. an irradiation temperature of 1108C before, during, and

after the irradiation was necessary to improve tensile

mechanical properties;

4. A2 samples exhibited the best mechanical performance

among those in the air irradiated samples;

5. V2 samples exhibited the best mechanical performance

among all the irradiated samples;

6. Average Et value of V2 samples increased about 40%,

average values of ry and ru increased both about 20%,

average eu value decreased about 22%, and average

work to fracture increased 5% compared to those of

UT samples.

TABLE II. Tensile Mechanical Parameters of the Untreated and All Irradiated UHMWPE Samples

Sample Code ET (MPa) ry (MPa) ru (MPa) eu (%) Work to Fracture (MPa)

UT 328 6 20 25 6 1 54 6 3 769 6 46 241 6 11

A1 493 6 45 23 6 1 48 6 1 485 6 42 155 6 13

A2 448 6 30 22 6 1 60 6 3 647 6 33 200 6 15

A3 430 6 18 22 6 1 47 6 1 595 6 53 172 6 13

V1 436 6 24 27 6 1 58 6 3 534 6 26 198 6 15

V2 460 6 26 30 6 2 65 6 4 595 6 32 254 6 9

V3 460 6 43 22 6 1 50 6 3 540 6 51 170 6 15

V4 355 6 19 25 6 1 55 6 3 634 6 27 208 6 12

V5 347 6 21 22 6 1 43 6 2 463 6 30 130 6 10

Figure 2. Stress–strain curves of UT, A1, A2, A3 samples (a) and ofUT, V1, V2, V3 samples (b).

58 VISCO ET AL.


A second mechanical comparison was made between the

best in vacuum irradiated sample (V2) with V4 and V5

one, in order to investigate about the irradiation tempera-

ture effects. Results clearly suggested that mechanical per-

formances of V4 and V5 samples were lower than that of

the V2 one. This confirmed that the temperature of 1108Cbefore, during, and after the in vacuum irradiation was suit-

able to obtain the best mechanical performance.

In air and in vacuum irradiated samples (at Tb, Ti, andTa of 1108C) were subjected to wear rate and hardness

measurements, in comparison to the UT sample. These

samples were irradiated with different doses, within the 0–

150 kGy range [Figure 3(a,b)].

Results showed that wear rate decreased with increasing

irradiation dose, both in air and in vacuum. When the irradia-

tion was performed in vacuum, a greater decrease in average

wear rate was observed, 73.8 and of 82.5% for the sample

irradiated in vacuum at 100 and 150 kGy, respectively, com-

pared to the UT samples. Also, an average decrease in wear

rate of 45.4 and of 47.6% was measured for samples irradi-

ated in air at 100 and 150 kGy, respectively, compared to the

UT samples. Hardness value of the UT sample increased as

the irradiation dose increased in air and in vacuum. When

the irradiation was performed in vacuum, hardness values

were lower. An average increase was observed of 1.93 and

6.98% for the sample irradiated in vacuum at 100 and

150 kGy, respectively, compared to the UT samples. Also,

an average increase of 16.41 and 22.67% was measured for

samples irradiated in air at 100 and 150 kGy, respectively,

compared to UT samples. So, we can conclude that A2 sam-

ple showed high stiffness and hardness, low deformability

but a poor wear resistance. On the contrary, the V2 sample

showed both good tensile and wear resistance, although its

hardness was lower than the A2 sample.

To connect the mechanical performance with the materi-

als structure induced by the electron beam irradiation pro-

cess, a physical investigation was performed by FTIR and

DSC analyses.

FTIR analysis was performed on of the UT, A1, A2,

A3, V1,V2, V3 samples. Qualitative results are shown in

Figures 4 and 5 and quantitative data are given in detail in

Table III. These last ones (RIvinyl, RItransvinylene, OICarbonyl,

OIHydroperoxides) indicate the preferential evolution of free

radicals produced by the electron beam irradiation. Vinyl

group reduction showed that chemical reaction occur in the

polymeric chains forming trans vinylene bonds, carbonyl

containing species (like ketones, esters, carboxylic acids),

and hydroxyl containing species (like hydroperoxides).18

The amount of these species strictly depends on the oxygen

presence beside the temperature before, during, and after

irradiation.

Our results showed that the vinyl group (CH2¼¼CH��peak at 908 cm21) of the UT sample was reduced after

electron beam irradiation, both in air and in vacuum in a

similar way [Figure 4(a,b) and Table III]. On the contrary,

the UT sample did not show the presence of a peak relative

to the trans vinylene group (��CH¼¼CH�� at 965 cm21)

that was instead present in all the other samples [Figure

4(c,d)]. Among all the investigated groups, the highest

RItransvinylene value was obtained in the A1 samples and the

lowest in the V3 samples. The irradiation process induced

the formation of C¼¼O [Figure 5(a,b)] ��O��OH species

[Figure 5(c,d)] in the irradiated samples. The curves of Fig-

ure 5(c,d) were normalized with respect to that of UT sam-

ple, where oxidized species are absent.19 Results suggested

that a clinically relevant oxidation occurred during the elec-

tron beam irradiation in air, especially in the A2 and A3

samples. Also, the C¼¼O species were formed in the V1

and V3 samples. Oxygen remained entrapped into the low

mobile chains of V1 samples and radicals could react with

it. Free radicals of V3 sample could react with oxygen after

irradiation, since they are not stabilized by thermal anneal-

ing. Oxidized species were absent in the V2 samples.

Among all the investigated specimens, the highest

OICarbonyl and OIHydroperoxides amount were obtained in the

A2 samples and the lowest in the V2 samples.

The crystallinity (vc) value of UT sample was approxi-

mately 52% (Table IV). This value increased up to about

54–60% in the A-series samples and up to about 69–81%

in the V-series samples. The highest crystallinity of 81%

Figure 3. Wear rate (a) and Vickers hardness (b) versus theadsorbed dose of UT and irradiated UHMWPE samples.



was reached in the V2 sample. Melting temperature (Tm)and lamellar thickness (Lc) values are given in detail in

Table IV. These values increased after the irradiation, com-

pared to that of the UT sample, regardless of the processing

environment. Additionally, melting temperature and lamel-

lar thickness values were higher in the A2, A3, V2, V3

samples compared with the A1 and V1 samples.

DISCUSSION

In this article, we studied the structural and mechanical

modifications induced by electron beam irradiation under

vacuum and in air on UHMWPE.

Some authors have already studied electron beam irradi-

ation in the presence of air,2 nitrogen,20 or in vacuum.21

They used different irradiation conditions to avoid exces-

sive loss of fracture and fatigue resistance and only

enhance its wear resistance. As a matter of fact it is hard to

compare the mechanical results obtained in this work with

those of other authors because of the different condition

and different aspects that have been investigated. Besides,

in this experimental research, a better vacuum (of 5 31026 mbar) was obtained compared with other authors

(that was 1024 and 1025 mbar).

Electron beam irradiation modifies UHMWPE structure

and, consequently, its mechanical performance in a differ-

ent way that depends on:

1. oxygen presence or absence during and after the irradi-

ation,

2. thermal heating before and during irradiation,

3. thermal annealing after irradiation.

Figure 4. FTIR analysis of UT, A1, A2, A3 samples and of UT, V1, V2, V3 samples within the 895–

925 cm21 range (a, b) and within the 940–990 cm21 range (c, d).

60 VISCO ET AL.


Irradiation produces free radicals in the polymer, both in

air and in vacuum. At room temperature (258C), free radi-

cals preferentially react with those close to them, into the

same chain, giving rise to double bond formation (intermo-

lecular reaction). A higher temperature (like 1108C) enhan-ces chains mobility so that free radicals can react with

radicals of other chains (intramolecular reaction), besides

those close to them. Intramolecular reactions form cross-

links between polymeric chains. In the absence of air, radi-

cals react with one another rather than with oxygen.

Crosslinks and double bonds preferentially occur. In the

presence of air, radicals react also with oxygen, forming

oxidized species.

A polymer which contains free radicals is not stable

since they tend to react in order to stabilize themselves.

Thermal annealing after irradiation is necessary to favor

free radical reactions and material stabilization. Free radi-

cals and/or oxidized species suggest the presence of broken

Figure 5. FTIR analysis of UT, A1, A2, A3 samples and of UT, V1, V2, V3 samples within the 1680–

1750 cm21 range (a, b) and of A1, A2, A3 samples and of V1, V2, V3 samples within the 3520–

3580 cm21 range (c, d).

TABLE III. Reaction Indices and Oxidation Indices Values of theUT, A1, A2, A3, V1, V2, V3 Samples

Sample Code RIVinyl RITransvinylene OICarbonyl OIHydroperoxide

UT 0.017 0.000 0.000 0.000

A1 0.006 0.082 0.208 0.006

A2 0.006 0.079 0.359 0.017

A3 0.006 0.071 0.349 0.016

V1 0.008 0.070 0.023 0.000

V2 0.006 0.072 0.000 0.000

V3 0.006 0.069 0.045 0.000



macromolecular chains. In contrast, the presence of cross-

links induces a stable network in the polymer.

The reduced chains mobility of the A1 and V1 samples

favored the formation of intramolecular double bonds de-

spite the intermolecular crosslinks.

The A3 and V3 samples, which contain free radicals,

showed the lowest amount of crosslinks and/or new chemi-

cal bonds (like trans vinylene double bounds or oxidized

species).

The A2 samples had a high amount of crosslinks (besides

trans vinylene double bounds and oxidized species) as sug-

gested by the appreciable increase in crystallinity.

The V2 samples showed the highest structural order and

therefore, the highest crosslink amount, with double bonds

presences and without any oxidized species.

So a thermal treatment at 1108C before, during, and af-

ter the irradiation, enhanced the molecular mobility favor-

ing free radicals reactions and their stabilization. At this

temperature the polymer chains are not yet melted but, due

to their gel state, they have enough mobility. In this way,

they create new molecular organizations. A temperature

lower than 1108C (258C, for example) does not allow the

chain fragments mobility. A temperature higher than 1108C(for example at 130 or 1508C) starts the process of material

melting, strongly decreasing its crystallinity and, hence, the

material stiffness.22

Moreover, UHMWPE irradiated at 1108C (before, dur-

ing and after) showed the best mechanical performance

among all the investigated samples. The ductile to brittle

transition of irradiated UHMWPE is partially reduced by

the crosslinks and double bonds formation beside the ab-

sence of oxygenated species. In fact the UHMWPE irradi-

ated in vacuum (V2) showed a better mechanical

performance compared with the air irradiated specimens

(A2).

This different behavior can be related to a different poly-

meric structure modification of both the crystalline ordered

and the amorphous disordered regions.

In the crystalline region, a structural order increase

occurred in all the samples after the radiation process,

regardless of the environmental conditions (air or vacuum),

as indicated by the DSC results. This increase is due to an

intermolecular bond formation in the amorphous phase

(such as crosslinks) and/or to crystalline lamellar thickness

growth in the ordered crystalline phase.

A thermal treatment at 1108C before and during the irra-

diation favors the crystals growth, but does not allow any

new crystal organizations. This organization is instead pos-

sible during the thermal annealing performed after irradia-

tion. In fact, during the 60 min at 1108C, free radicals

undergo molecular reorganizations. Crosslinks are created

and new crystal organizations are enhanced, according to

Medel et al.23 For these reasons the A3 and V3 samples,

which were heated only before and during irradiation,

showed the lowest crystallinity and therefore, the lowest

crosslink amount among the studied samples. Besides, their

lamellar thickness is higher than that of A1 and V1 but

comparable to that of A2 and V2 one. Also the A1 and V1

samples, which were heated only after irradiation, had the

smallest lamellar thickness and a crystallinity higher than

that of A3 and V3 samples. Additionally, the A2 and V2

samples showed the highest lamellar thickness and the

highest crystallinity since they contained the highest

amount of crosslinks among the studied samples.

So, a thermal treatment before, during, and after the irra-

diation is necessary to increase both the crystallinity and

their lamellar thickness. These microstructural parameters

influence the mechanical performance of the UHMWPE.

Among the investigated physical–mechanical parame-

ters, it was observed that only the material stiffness

increase in the UHMWPE regardless of the radiation envi-

ronment. All the other parameters were instead influenced

by the presence of air during the radiation process. For this

reason, an increase in crystallinity could be related to the

enhancement in material stiffness.24 Electron beam radia-

tion modifies the disordered amorphous regions, producing

both crosslinks and oxidized species depending on the irra-

diation environment.25 Crosslink preferentially occurs in

vacuum while oxidized species preferentially occurs in air,

as suggested by the FTIR and DSC results.

Wear measurements also confirmed that a high amount

of crosslinks was formed during the in vacuum irradiation.

The presence of crosslinks prevents free sliding of the

polymer chains; debris formation can be highly contrasted

by the reinforced chains which better resist the rotating pin

mechanical stress. The reinforcement action of the cross-

links enhance the polymeric mechanical wear resistance.14

So, the samples irradiated in air (A2) had lower wear re-

sistance than samples irradiated in vacuum, (V2). On the

contrary, A2 showed a higher hardness compared to V2

and UT samples. A possible explanation for the different

hardness could be related to the chemical modification of

amorphous chains during irradiation in air. The formation

of new oxidized species probably induces new hydrogen

bonds or, generally, new physical intermolecular interac-

tions. These interactions harden the macromolecular struc-

ture of A2 samples that became stronger, less deformable,

and more brittle compared to UT and V2 samples.

TABLE IV. Crystallinity, Melting Temperature, and LamellarThickness Values of the UT, A1, A2, A3, V1, V2, V3 Samples

Sample

Code

Crystallinity

vc (%)

Melting

Temperature

Tm (8C)

Lamellar

Thickness

LC (nm)

UT 52.7 6 0.7 135.9 6 0.5 27–30

A1 56.5 6 1.2 137.9 6 0.7 35–40

A2 59.6 6 1.8 138.6 6 0.4 37–42

A3 54.2 6 1.1 138.7 6 0.5 37–43

V1 71.9 6 1.9 137.6 6 0.8 32–39

V2 81.1 6 2.2 138.4 6 0.6 35–42

V3 68.8 6 2.0 138.3 6 0.7 34–41

62 VISCO ET AL.


The different yielding strength could be related to the

changes in molecular mobility that occurred in the amor-

phous region depending on the irradiation environment.25

In the presence of oxygen, as in the presence of free radi-

cals, chain scission occurs in the amorphous region. The

broken chains are easily orientated toward the stress direc-

tion favoring polymeric yielding. For this reason, the A-se-

ries samples yielding stress was lower compared with the

V-series samples ( except V3 that contained free radicals)

and untreated UHMWPE samples.

The yield stress of polyethylene increases with crystal

thickness up to 40 nm and saturates above this value, as

reported by Galeski.26 The crystal thickness of �40 nm,

together with the oxygen absence and a high crosslink

amount, explains the highest yield stress of the V2 samples

among all the studied samples.

These observations suggested that heating during the

irradiation with electron beam and annealing after irradia-

tion are useful to improve the material only if oxygen is

eliminated by vacuum and if free radical are stabilized by

an annealing treatment. In fact, in vacuum the material

stiffness, work to fracture, strength, tensile yielding, and

wear resistance increased.

So, experimental results showed that high vacuum irra-

diation at 1108C and a post-treatment at 1108C for 1 h

gives the highest material mechanical performance for all

UHMWPEs studied.

This observed improvement in mechanical performance

was very attractive, however for medical applications of

UHMWPE components, the costs must also be considered,

since the high vacuum application is expensive. We should

also remember that changes in prostheses also require an

additional surgical operation. For this reason it is necessary

to carry out a more detailed investigation in order to choose

the very best UHMWPE treatment to increase in vivo per-

formance, and thereby decrease surgical failures. With this

aim, further investigations such as fatigue resistance will be

necessary to predict the material behavior during dynamic

stresses. In addition, it should also be interesting to perform

accelerate ageing in the presence of oxygen27 to check the

long-term stability of the modified material.

CONCLUSIONS

In this study, electron-beam irradiations were performed on

UHMWPE samples in air and in vacuum under different

thermal conditions. Among all the tested conditions, the

100 kGy irradiated UHMWPE under vacuum at 1108C and

annealed at the same temperature for 1 h exhibited good

mechanical performance and wear resistance.

Heating before and during irradiation at the temperature

of 1108C enhanced chain mobility and crystal growth. Dur-

ing the thermal annealing just after the irradiation, at the

same temperature, intramolecular double bonds and inter-

molecular crosslinks were highly favored. Free radicals

react in such a way to obtain a structurally stable material.

The electron beam irradiation process changed the mate-

rial structure and, consequently, its mechanical perform-

ance. Ordinate crystalline region increased and stiffened

the polymer. Amorphous region changes are related to oxy-

gen presence and thermal annealing. The oxygen presence

and the lack of thermal annealing enhanced chains scis-

sions reaction compared with the crosslinks and double

bonds formation. Vice versa in vacuum and with thermal

annealing. When chain scission occurred, the broken chains

were easily orientated toward the stress direction. This

favored the material yielding at lower stress values with

respect to the untreated UHMWPE. The oxidized species

presence hardened the material, which became brittle for

the physical intermolecular interactions between the chains.

In vacuum, oxygen presence was highly reduced. The

materials showed a high wear resistance and tensile yielding

since the high amount of crosslinks bonds stiffened the

amorphous chains opposing to the tensile and wear stresses.

REFERENCES

1. Costa L, Jacobson K, Bracco P, Brach del Prever EM. Oxida-tion of orthopaedic UHMWPE. Biomaterials 2002;23:1613–1624.

2. Oral E, Malhi AS, Muratoglu OK. Mechanisms of decrease infatigue crack propagation resistance in irradiated and meltedUHMWPE. Biomaterials 2006;27:917–925.

3. Yamamoto K, Masaoka T, Manaka M, Oonishi H, Clarke I,Shoji H, Kawanabe K, Imakiire A. Micro-wear features onunique 100-Mrad cups: Two retrieved cups compared to hipsimulator wear study. Acta Orthop Scand 2004;75:134–141.

4. Zhu C, Jacobs O, Jaskulka R, Koller W, Wu W. Effect ofcounterpart material and water lubrication on the sliding wearperformance of crosslinked and non-crosslinked ultra high mo-lecular weight polyethylene. Polym Testing 2004;23:665–673.

5. Oral E, Wannomae KK, Hawkins N, Harris WH, MuratogluOK. a-Tocopherol-doped irradiated UHMWPE for high fa-tigue resistance and low wear. Biomaterials 2004;25:5515–5522.

6. Muratoglu OK, Delaney J, O’Connor DO, Harris WH. Theuse of trans-vinylene formation in quantifying the spatial dis-tribution of electron beam penetration in polyethylene. Single-sided, double-sided and shielded irradiation. Biomaterials 2003;24:2021–2029.

7. Edidin AA, Jewett CW, Kalinowski A, Kwarteng K, KurtzSM. Degradation of mechanical behavior in UHMWPE afternatural and accelerated aging. Biomaterials 2000;21:1451–1460.

8. Muratoglu OK, Bragdon CR, O’Connor DO, Jasty M, HarrisWH, Gul R, McGarry F. Unified wear model for highly cross-linked ultra-high molecular weight polyethylenes (UHMWPE).Biomaterials 1999;20:1463–1470.

9. Torrisi L, Visco AM, Barna R, De Pasquale D, Campo N, DiMarco G, Trimarchi M, Trifiro A. Radiation effects inducedby MeV electron beams irradiating dense polyethylene(UHMWPE). Radiat Eff Defects Solids 2004;159:259–271.

10. Currier BH, Currier JH, Collier JP, Mayor MB, Scott RD.Shelf life and in vivo duration. Impacts on performance oftibial bearings. Clin Orthop Relat Res 1997;342:111–122.

11. Goldman M, Gronsky R, Pruitt L. The influence of steriliza-tion technique and ageing on the structure and morphologyof medical-grade ultrahigh molecular weight polyethylene.J Mater Sci Mater Med 1998;9:207–212.



12. Nakamura K, Ogata S, Ikada Y. Assessment of heat and stor-age conditions on c-ray and electron beam irradiatedUHMWPE by electron spin resonance. Biomaterials 1998;19:2341–2346.

13. Suarez JCM, de Biasi RS. Effect of gamma irradiation on theductile-to-brittle transition in ultra-high molecular weightpolyethylene. Polym Degrad Stab 2003;82:221–227.

14. Valenza A, Visco AM, Torrisi L, Campo N. Characterizationof ultra-high-molecular-weight polyethylene (UHMWPE)modified by ion implantation. Polymer 2004;45:1707–1715.

15. Torrisi L, Visco AM, Campo N. Dose and dose-rate depend-ence of in air electron beam irradiation of polyethylene.J Mater Eng Perform 2007;16:97–101.

16. Solti A, Hummel DO, Simak P. Computer-supported infraredspectroscopy of polyethylene, ethylene copolymers, and amor-phous poly(alkyl)ethylenes. Makromol Chem Macromol Symp1986;5:105–133.

17. Medel FJ, Garcıa-Alvarez F, Gomez-Barrena E, Puertolas JA.Microstructure changes of extruded ultra high molecularweight polyethylene after gamma irradiation and shelf-aging.Polym Degrad Stab 2005;88:435–443.

18. Brunella V, Bracco P, Carpentieri I, Paganini MC, Zanetti M,Costa L. Lifetime of alkyl macroradicals in irradiated ultra-high molecular weight polyethylene. Polym Degrad Stab 2007;92:1498–1503.

19. Jacobson K. Oxidation of ultra high molecular weight poly-ethylene (UHMWPE), Part 1: Interpretation of the chemilumi-nescence curve recorded during thermal oxidation. PolymDegrad Stab 2006;91:2126–2132.

20. Muratoglu OK, O’Connor DO, Bragdon CR, Delaney J, JastyM, Harris WH, Merrill E, Venugopalan P. Gradient crosslink-ing of UHMWPE using irradiation in molten state for totaljoint arthroplasty. Biomaterials 2002;23:717–724.

21. Tretinnikov ON, Ogata S, Ikada Y. Surface crosslinking ofpolyethylene by electron beam irradiation in air. Polymer1998;39:6115–6120.

22. Kurtz SM, Villarraga ML, Herr MP, Bergstrom JS, Rimnac CM,Edidin AA. Thermomechanical behavior of virgin and highlycrosslinked ultra-high molecular weight polyethylene used intotal joint replacements. Biomaterials 2002;23:3681–3697.

23. Medel FJ, Pena P, Cegonino J, Gomez-Barrena E, PuertolasJA. Comparative fatigue behavior and toughness of remeltedand annealed highly crosslinked polyethylenes of electronbeam irradiated UHMWPE. J Biomed Mater Res Part B:Appl Biomater 2007;83:380–390.

24. Buchanan FJ, White JR, Sim B, Downes S. The influence ofgamma irradiation and aging on degradation mechanisms ofultra-high molecular weight polyethylene. J Mater Sci MaterMed 2001;12:29–37.

25. Bracco P, Brunella V, Luda MP, Zanetti M, Costa L. Radia-tion-induced crosslinking of UHMWPE in the presence of co-agents: Chemical and mechanical characterisation. Polymer2005;46:10648–10657.

26. Galeski A. Strength and toughness of crystalline polymer sys-tems. Prog Polym Sci 2003;28:1643–1699.

27. Burroughs BR, Blanchet TA. The effect of pre-irradiationvacuum storage on the oxidation and wear of radiation steri-lized UHMWPE. Wear 2006;261:1277–1284.

64 VISCO ET AL.


Mechanical performance of electron-beam-irradiated UHMWPE in vacuum and in air

Documents