lable at ScienceDirect

Polymer Degradation and Stability 96 (2011) 470e476

Contents lists avai

Polymer Degradation and Stability

journal homepage: www.elsevier .com/locate/polydegstab

Mechanical behaviour of a poydimethylsiloxane elastomer after outdoorweathering in two different weathering locations

Panagiota N. Eleni a,*, Magdalini K. Krokida a, Gregory L. Polyzois b, Constantinos A. Charitidis a,Elias P. Koumoulos a, Vasiliki P. Tsikourkitoudi a, Ioannis Ziomas a

a School of Chemical Engineering, National Technical University of Athens, Zografou Campus, 15780 Athens, GreecebDepartment of Prosthodontics, Division of Removable Prosthodontics, Dental School, University of Athens, 2 Thivon Str Goudi, 115 27 Athens, Greece

a r t i c l e i n f o

Article history:Received 20 July 2010Received in revised form5 January 2011Accepted 13 January 2011Available online 26 January 2011

Keywords:CompressionNanoindentation analysisNatural weatheringPolydimethylsiloxaneTensile testing

* Corresponding author. Tel.: þ30 2107723149; faxE-mail address: peleni@central.ntua.gr (P.N. Eleni)

0141-3910/$ e see front matter � 2011 Elsevier Ltd.doi:10.1016/j.polymdegradstab.2011.01.017

a b s t r a c t

The degradation of maxillofacial prosthetic elastomers that occur during physical weathering is usuallyresponsible for the replacement of the prosthesis. In this study the mechanical behaviour of a poly-dimethylsiloxane (PDMS) elastomer was investigated, after 1 year outdoor weathering in two differentweathering locations in Greece (Thessaloniki, Athens). The hypothesis investigated was that irradiationtime did not affect the measured properties. Specimens (Elastomer 42) were prepared according tomanufacturer’s instructions and exposed to solar radiation for 1 year. Compression, tensile and nano-indentation tests were performed before and after the exposure. Compression and tensile data were alsosubjected to analysis of variance (ANOVA) and Tukey Post hoc tests at a level of a ¼ .05. These propertieswere selected due to their clinical significance for fabrication and maintenance of a facial prosthesis.According to statistical analysis all the measured properties changed significantly after outdoorweathering. More specifically, most of the properties presented significant changes after six months ofweathering. The observed changes also depended on the weathering locations. The hypothesis investi-gated was rejected. Material A became harder and the observed differences in the mechanical behaviourresulted from photo-degradation and hydrolysis that might occur due to weathering. The study alsoprovides new information about maxillofacial prosthetics serviceability obtained from nanoindentationtests.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction

External maxillofacial prostheses are used to rehabilitateanatomy, function or cosmetics of facial regions that are missing orchanged due to disease, accident or congenital malformation, andcannot be surgically restored [1,2]. Different types of materials areused to fabricate maxillofacial prostheses including poly(methylmethacrylate), poly(vinyl chloride), chlorinated polyethylene,polyurethanes, and silicones [2e5]. Polymeric materials are highlyversatile, with low thermal stability and little resistance to solarradiation and therefore their performance is still far from ideal [6].Polydimethylsiloxanes are the most widely used materials inmaxillofacial prosthesis mainly due to their easy manipulation. Inaddition, comparing to other polymer prosthetics, they presentexcellent overall radiation resistance [7e11].

: þ30 2107723155..

All rights reserved.

Natural or outdoor weathering of polymers can induce signifi-cant changes in their chemical, physical, and mechanical proper-ties. The main climate characteristics that cause degradation aresunlight, temperature, moisture, wind, dust and pollutants. Moreprecisely, deterioration is a photo-oxidative attack; that is thecombined action of oxygen and sunlight on their chemical struc-ture. Thus, it is preferable to have factual information on the actuallong term performance of a material outdoors instead of an artifi-cial weathering. The destructive effect of the weather on polymericmaterials has strong dependence on geographic location, season,time of day, cloud cover and exposure orientation, since the criticalweather factors vary with these conditions. Different climateconditions can also be observed during different seasons and years,so it is preferable to run outdoor tests during different seasons andover a period of at least one year [12]. Artificial weathering can alsoapproximate the outdoor performance of polymers and in manycases is used to predict the lifetime of polymers under serviceconditions [13e15]. However, accelerated weathering might influ-ence the degradation mechanism and could lead to totally wrongestimates of the lifetime of polymers [13,16,17]. In previous studies

Table 2Monthly average radiation and climatic data during outdoor weathering.

Date Radiation and climatic data

T (�C) Rain (mm) TSR (kW/m2)

May 2007 20.65 56.80 0.22June 2007 26.10 42.60 0.22July 2007 28.20 0.00 0.33August 2007 27.00 53.20 0.27September 2007 21.20 34.80 0.18October 2007 17.10 64.40 0.13November 2007 11.10 49.00 0.08

P.N. Eleni et al. / Polymer Degradation and Stability 96 (2011) 470e476 471

mathematical modelling has been presented as an essential processin order to predict the measured properties with solar or acceler-ated irradiation time [18e21].

The aim of this study was to investigate the effect of irradiationtime and weathering location after aging for 1 year on somemechanical properties i.e. tensile and compress strength, modulusof elasticity, elongation at break, and nano-hardness, of a maxillo-facial PDMS elastomer. Our null hypothesis stated that the exam-ined mechanical properties of maxillofacial silicone elastomer arenot affected by irradiation time and weathering location.

December 2007 6.70 15.00 0.06January 2008 6.70 25.80 0.08February 2008 8.40 20.8 0.12March 2008 12.90 14.00 0.14April 2008 15.00 73.10 0.20May 2008 19.60 25.70 0.29

T¼ Temperature, Rain¼Monthly summative rain height, TSR¼ total solar radiation,Data source: Laboratory of Atmospheric Physics of Aristotle University of Thessa-loniki, personal communication, 2007e2008.

2. Materials and methods

2.1. Materials

The commercial material used and the curing conditionsemployed in this study are shown in Table 1. Forty two rectangularspecimens (15� 20� 35 mm) for nanomechanical analysis, andforty two dumbbell-shaped type II for tensile tests were fabricated.Six specimens from each shape were considered as control.

2.2. Outdoor weathering

Outdoor weathering experiments were performed in twodifferent weathering locations in Greece, in Athens and Thessalo-niki. Specimens were placed on the roof of the laboratory ofatmospheric physics of Aristotle University of Thessaloniki fromMay 2007 through May 2008 and on the roof of the school ofChemical Engineering of National Technical University of Athensfrom July 2007 through July 2008. The monthly average radiationand climatic data in Thessaloniki and Athens during outdoorweathering are presented in Tables 2 and 3, respectively. During theweathering, specimens were left uncovered and exposed to theenvironmental conditions. The exposure rack was adjusted to anangle of 5� from the horizontal to avoid standing water andmaximize the amount of sunlight on the specimens. Every twomonths 3 rectangular and 3 dumbbell-shaped specimens from eachlocation were analyzed. Before specimens were measured, theywere cleaned for 10 min in distilled water in an ultrasonic cleaner,dried to a constant weight in a desiccator to an accuracy of 0.001 g(Kern EW balance, Kern & Sohn GmbH, Ziegelei, 72336 Baling-enand) and tested. First compression was conducted and at thesixth and twelfth month after compression nanoindentation anal-ysis was also performed. The dumbbell-shaped specimens wereplaced in tension in order to perform tensile tests.

2.3. Compression test

Compression tests were conducted using a Universal TestingMachine Zwick model Z2.5/TN1S (Germany). The uniaxialcompression tests were performed at room temperature (25 �C).Constant deformation rate was set at 5 mm/min for all examinedmaterials. Force and deformation were recorded electronically andthe resulting stressestrain compression curves were constructed.

Table 1Materials of the study.

Coding Material Type Man

A Elastomer42

Addition (platinum) type-RTV(room temperature vulcanizing)

TechNew

2.4. Tensile test

The dumbbell-shaped specimens were placed in tension ina Universal Testing Machine Zwick model Z2.5/TN1S (Germany)that was supplied with automatic extensometer. The extensometergrips were set to a standard length of 20 mm, and the crossheadspeed was set at 5 mm/min for all examinedmaterials. The uniaxialtensile tests were performed at room temperature (25 �C). Forceand elongationmeasurements were recorded electronically and theresulting stressestrain tensile curves were constructed.

2.5. Nanoindentation analysis

The nanoindentation analysis was performed using a HysitronTriboLab� (Brazil) Nanomechanical. Test instrument allowing theapplication of loads from 1 to 10.000 mN and the recording ofpenetration depths as a function of applied loads with a high loadresolution (1 nN) and a high displacement resolution (0.04 nm).The TriboLab� employed in this study was equipped with a Scan-ning Probe Microscope (SPM), in which a sharp probe tip moved ina raster scan pattern across a sample surface using a three-axispiezo positioner. In all depth-sensing tests a total of 10 indentswere averaged to determine the mean values of nano-Hardness (H)and modulus of Elasticity (E) for statistical purposes, with a spacingof 50 mm (w45% relative humidity, 23 �C).

2.6. Mathematical modelling

Several mathematical models were used in order to predict thedependence of stress, strain, elasticity and viscoelasticity param-eter on irradiation time for both compression and tensile tests.

Mathematical models were fitted to the experimental data andthen to statistical analysis of variance (ANOVA). Mathematicalmodels which, according to statistical analysis, present the lowerstandard deviation value from the experimental datawere selected.Thus the selectedmathematical models that were themost suitable

ufacturer Mixing ratio Curing method

novent Ltd, Principality House,port, South Wales, UK

(10:1) Curing at 100 �Cfor 2 h

Table 5Mathematical model for tensile tests.

Tensile test

Viscoelastic behaviour (stressestrain equation)st ¼ Et*3t þ ðsmax;t � Et*3max;tÞ*ð3t=3max;tÞpt

Parameterssmax,t e maximum stress (MPa)3max,t e maximum strain (mm/mm)Et e elasticity parameter (MPa)pt e viscoelasticity parameter (e)

Parameter equationssmax;t ¼ s0;t þ s1;t*ðtir=t0Þk53max;t ¼ 30;t þ 31;t*ðtir=t0Þk6Et ¼ E0;t þ E1;t*ðtir=t0Þk7pt ¼ p0;t þ p1;t*ðtir=t0Þk8wheretir e irradiation time (h)t0 e reference time (h)

Table 3Monthly average radiation and climatic data during outdoor weathering.

Date Radiation and climatic data

T (�C) Rain (mm) TSR (kW/m2)

July 2007 29.02 0.0 0.33August 2007 27.90 16.2 0.29September 2007 22.56 0.0 0.24October 2007 18.13 79.2 0.16November 2007 12.99 34.6 0.09December 2007 8.41 54.2 0.08January 2008 7.42 27.0 0.09February 2008 7.76 31.6 0.14March 2008 11.17 55.0 0.20April 2008 15.64 62.0 0.23May 2008 19.73 2.2 0.30June 2008 25.76 1.2 0.34July 2008 27.78 0.0 0.33

T¼ Temperature, RH¼ Relative humidity, TSR¼ Total solar radiation, Data source:METEONET (http://meteonet.chi.civil.ntua.gr/gr/divs.html), Laboratory ofHydrology and Resources Management, NTUA, 2007e2008.

0

1

2

3

10

Stre

ss (M

Pa)

Strain (mm/mm)

compress dataCALCULATED

a

P.N. Eleni et al. / Polymer Degradation and Stability 96 (2011) 470e476472

and simple, referring to compression and tensile data, aresummarized in Tables 4 and 5, respectively [22,23].

The stressestrain equation (Table 4) that describes the visco-elastic behaviour (Fig. 1a) involves four parameters: the maximumstress (smax), the corresponding strain (3max), the modulus ofelasticity (E), and the viscoelasticity parameter (p). Maximum stressand strain represent the maximum point to which the samples canbe compressed. The modulus of elasticity represents the linearelastic behaviour part of the stressestrain curve and shows theelastic nature of the material. The viscoelasticity parameter repre-sents the exponential part of the curve.

Tensile data were fitted in the described equations in Table 5.The examined silicone elastomer (PDMS-material A) is brittle anddoes not indicate yield point on its tensile behaviour (Fig. 1b). Thestressestrain equation (Table 5) that describes the viscoelasticbehaviour involves four parameters: smax, 3max, E, and p.

2.7. Statistical analysis

Compression and tensile data were subjected to statisticalanalysis. All data were first evaluated for homogeneity of variancesby Levene’s test and KolmogoroveSmirnov for normality. Two-wayanalysis of variance was used to evaluate any differences betweenspecimens subjected to weathering in Athens and Thessaloniki, inorder to detect significant differences among the two weatheringlocations, the time intervals and possible interaction effects

Table 4Mathematical model for compression tests.

Compression test

Viscoelastic behaviour (stressestrain equation)sc ¼ Ec*3c þ ðsmax;c � Ec*3max;cÞ*ð3c=3max;cÞpc

Parameterssmax,c e maximum stress (MPa)3max,c e maximum strain (mm/mm)Ec e elasticity parameter (MPa)pc e viscoelasticity parameter (e)

Parameter equationssmax;c ¼ s0;c þ s1;c*ðtir=t0Þk1;c3max;c ¼ 30;c þ 31;c*ðtir=t0Þk2;cEc ¼ E0;c þ E1;c*ðtir=t0Þk3;cpc ¼ p0;c þ p1;c*ðtir=t0Þk4;cwheretir e irradiation time (h)t0 e reference time (h)

between these two factors. In addition, an overall one way ANOVAand Tukey post hoc tests were applied to detect differences amongall the groups for the properties that two-way analysis presentedsignificant interaction effects between two factors. All analyseswere computed with the SPSS for Windows software (SPSS 16.0,SPSS Inc, Chicago, Ill.) at a significance level of a¼ .05.

0

1

2

3

210

Stre

ss (M

Pa

)

Strain (mm/mm)

tensile dataCALCULATED

b

Fig. 1. Typical curves for (a) compression and (b) tensile analysis.

P.N. Eleni et al. / Polymer Degradation and Stability 96 (2011) 470e476 473

3. Results

The typical curves obtained from compress and tensile tests arepresented in Fig. 1a and b, respectively. The mathematical modelswere fitted to the experimental data. The correlation of smax, 3max, Eand p of compression and tensile analysis with irradiation time ispresented in Fig. 2. Figures clearly illustrate thatmaterial A presentsa decrement in smax and 3max for both weathering locations. Inaddition, elasticity and viscoelasticity parameters were increasedas irradiation time increased.

The values of the parameters involved in mathematical models,for compression and tensile tests, were calculated using equationsin Tables 4 and 5, respectively. The mathematical models’ param-eters estimation results are summarized in Tables 6 and 7 forcompression and tensile analysis, respectively.

A typical nanoindentation test provides loadedisplacementdata, which are the deformation responses of a material. In general,traditional mechanical property parameters, such as nano-Hardness (H) and modulus of Elasticity (E), can be determined.Attention had been paid on analyzing the unloading curve (first 30%of the unloading data) to obtain contact area for the determinationof H and E. Most analyses are based on the OliverePharr (O&P)method [24], which determines the contact area in the use of theunloading tangent (green line, Fig. 3) together with the known areafunction. In Fig. 3, typical loadeunload curves on samples i) controland ii) after 1 year weathering in Athens are presented. As can beseen in Fig. 3, for the same displacement (1700 nm), a controlsample revealed higher H, E values in comparison to those of 1 yearAthens sample. PDMS control sample showed greater resistance(enhanced nanomechanical properties), whereas the degradation(surface-bulk degradation) of 1 year physically aged PDMS (1 yearAthens) is confirmed. (For interpretation of the references to colourin Fig. 3, the reader is referred to the web version of this article.)

Adhesion between the tip and the sample can interfere withmeasurements of modulus using the compliance method in

Fig. 2. Maximum stress, maximum strain, elasticity parameter and viscoelasticity p

polymeric and soft tissue samples [25e28]. Adhesion is observed ina loadedisplacement curve as a region of negative load duringunloading. Recent studies of soft polymers (silicones) havedemonstrated that the compliance method overestimates thesample modulus when there is significant tip-sample adhesion[25,28]

The surface degradation of PDMS (cross-linking density,molecular weight, polar/non-polar molecules) caused by hydro-lysis, irradiation and oxidation had mainly affected the hydropho-bicity of the surface resulting lower adhesion with the tip [29,30].

Statistical analysis indicates significant changes due to irradia-tion time and weathering location. In addition, an interaction oftime intervals and weathering locations were also detected for Ec,3max,t, Et and pt. The results are summarized in Table 8. All themeasured properties changed significantly while irradiation timewas increased. More specifically for samples irradiated in Athens,smax,c and pc changed significantly after eight months, 3max,c and Ecafter two months and pc, smax,t, 3max,t, Et and pt after four months.Moreover, for samples whichwere irradiated in Thessaloniki, smax,c,3max,c, st, and 3max,t, changed significantly after two months, Ec aftersix months and pc, Et and pt after eight months. For all the prop-erties except pt significant differences were detected between thetwo weathering locations. Moreover, 3max,c, Ec, Et, smax,t and pt weresignificantly different concerning the weathering location. Greaterchanges were detected in samples in Athens except the st whichpresented greater changes in samples in Thessaloniki. Post hocanalyses presented similar behaviour for Ec, Et and pt in Thessalo-niki and Athens at the first two and the fifth time interval (second,fourth and tenth month).

4. Discussion

The null hypothesis was rejected, since according to statisticalanalysis significant changes in mechanical properties wereobserved.

arameter of compression and tensile analysis correlated with irradiation time.

Table 6Parameter estimation for compression model.

Weathering location s0,c (MPa) s1,c (MPa) 30,c (mm/mm) 31,c (mm/mm) E0,c (MPa) E1,c (MPa) p0,c (e) p1,c (e) k1 (e) k2 (e) k3 (e) k4 (e)

Athens 2.10 �0.39 2.56 �0.53 0.69 0.59 1.30 0.80 0.49 1.40 0.63 1.04Thessaloniki 2.22 �0.81 2.56 �0.94 0.70 0.23 1.26 0.38 0.22 0.58 1.39 1.55

Table 7Parameter estimation for tensile model.

Weathering location s0,t (MPa) s1,t (MPa) 30,t (mm/mm) 31,t (mm/mm) E0,t (MPa) E1,t (MPa) p0,t (e) p1,t (e) k5 (e) k6 (e) k7 (e) k8 (e)

Athens 2.69 �0.13 0.75 �0.26 0.27 1.15 2.50 0.27 0.46 0.35 0.50 1.50Thessaloniki 2.65 �0.26 0.73 �0.16 0.27 0.45 2.47 0.25 0.13 0.17 1.38 1.52

P.N. Eleni et al. / Polymer Degradation and Stability 96 (2011) 470e476474

Maxillofacial materials’ hardness is a measure of flexibility.Tensile strength is important to express overall strength charac-teristics. The resulting strain (elongation) is a measure of flexibilityand an indicator of the overall flexibility of a prosthetic materialwith facial movement. It also defines the resistance of a facialprosthetic elastomer to rupture during use and maintenance. Thisproperty also defines the material’s ability to accommodate facialmovement [31]. Relationships between physical structure andmechanical properties allow the understanding of mechanicalbehaviour around working temperature as well as the physicalaging [8].

The weathering of polymers leads to changes in physical andchemical characteristics that cause a significant deterioration inimportant mechanical properties. When a photo-oxidative degra-dation occurs the following steps can be considered:

1. Initiation Step: Free radicals, such as HO*2, P* (free polymerradical) and *OH are formed due to the presence of air (oxygen)under UV/VIS irradiation.

2. Propagation Step: The reaction of free polymer radicals withoxygen, leads to a production of polymer oxy and peroxy-radicals and secondary polymer radicals, which result in chainscission. The propagation step is very much dependent on theefficiency of the decomposition, photolysis and/or thermolysisof polymer hydroperoxides (POOH), during which new freeradicals, such as polymer oxy radical (PO*) and hydroxyl radical(HO*), are formed.

-1000 -500 0 500 1000 1500 2000-5

0

5

10

15

20

25

30

Load

(μΝ

)

Displacement (nm)

PDMS Control 1 year Athens

Hardness Elastic modulus

PDMS Control 0,79 MPa 5,59 MPa

1 year Athens 0,41 MPa 3,86 MPa

1700

Fig. 3. Typical loadeunload curves obtained from nanoindentation testing for pristine(material A Control) and 1 year physically aged (material A after weathering 1 year inAthens).

3. Termination Step: The termination of the polymer radicalsoccurs mainly by bimolecular recombination, but also betweenlow molecular radicals, such as hydroxyl (OH*) and hydro-peroxy (HO*2) and other available radicals (R*). There areseveral factors that influence the recombination reactions, suchas concentration of radicals formed and structural parametersof the polymeric matrix (free volumes). Moreover, oxygenpressure controls which of the reactions overrules. For examplewhen the oxygen pressure is high (atmospheric pressure), thetermination reaction almost exclusively occurs by bimolecularreaction of PO*. The reaction of different free radicals with eachother resulting in cross-linking [32].

The main structural modifications in irradiated polymers arechanges in molecular weight distribution e due to main chainscission, cross-linking and end linking e and the production ofvolatile degradation products [33e35]. The main volatile productsof irradiated polydimethylsiloxanes are hydrogen, methane andethane gases [36]. All these phenomena tend to modify the mate-rials’ physical properties.

The changes of physical properties affect the polymer’s struc-tural network in different ways. The structural networks densityincreases during cross-linking, due to the formation of bondsbetween the existing chain segments or between the chains.Therefore, cross-linking leads to harder materials. On the otherhand, when chain scission is the dominant mechanism, the frac-turing bonds within the main chain or between two differentchains, incur a decrement in density of the structural network andthe materials become softer. In irradiated polymers, both the abovemechanisms take place.

Over the past 50 years a solid understanding of the relationshipbetween polymer structure and the relative yields of cross-linkingand chain scission has been acquired. Thus, NMR and IR experi-ments show that polydimethylsiloxanes, undergo mostly primarilycross-linking reactions [36].

Elasticity and strength are highly dependent on the molecularweight and the degree of cross-linking of a polymer. Flory mathe-matical model correlates polymers’ network density with E(Equation (1)) [37].

E ¼ RTr

Mc

�3þ 2

32

�(1)

where Mc is the molecular weight between cross-links and r

symbolizes the network density.The correlation between Elastic modulus and molecular weight

(Mc), which was determined from swelling measurements, isillustrated in Fig. 4.

The comparison of modulus (E) values measured using nano-indentation to handbook or manufacturer values can often be

Table 8Statistical analysis and significant changes.

smax,c 3max,c Ec pc smax,t 3max,t Et pt

Weathering location p< 0.05 p< 0.05 p< 0.05 p< 0.05 p< 0.05 p< 0.05 p< 0.05 p> 0.05Irradiation time p< 0.05 p< 0.05 p< 0.05 p< 0.05 p< 0.05 p< 0.05 p< 0.05 p< 0.05Interaction (weathering location * Irradiation time) p> 0.05 p> 0.05 p< 0.05 p> 0.05 p> 0.05 p< 0.05 p< 0.05 p< 0.05

0

4000

8000

12000

16000

0 0.5 1 1.5 2

Mo

lecu

lar w

eig

ht b

etw

een

cro

sslin

ks

(g

/m

ol)

Modulus of Elasticity (MPa)

A_Athens_tensile_12 monthsA_Thessaloniki_tensile_12 monthsA_control

Fig. 4. Elastic modulus correlated with molecular weight between cross-links.

0

1

2

3

4

5

6

A_Control A_Thessaloniki A_Athens

Co

mp

re

ss

mo

du

lu

s (M

Pa)

Samples

E_Oliver-PharrE_Compress

In

de

nta

tio

n

Ela

stic

mo

du

lu

s (M

Pa)

0

1

2

3

4

5

6

A_Control A_Thessaloniki A_Athens

Te

ns

ile

mo

du

lu

s (M

Pa)

Samples

E_Oliver-PharrE_Tensile

In

de

nta

tio

n

Ela

stic

m

od

ulu

s

(M

Pa

)

a

b

Fig. 5. Comparison of elastic moduli obtained from nanoindentation (using theOliverePharr model), (a) compressive and (b) tensile testing.

P.N. Eleni et al. / Polymer Degradation and Stability 96 (2011) 470e476 475

misleading, because quoted values of E for many polymer systemscan cover a large range due to potential variations in microstruc-ture, semicrystalline morphology, anisotropy, molecular weight,cross-link density, etc. The comparison of elastic moduli obtainedfrom nanoindentation (using the OliverePharr model), tensile andcompressive testing is illustrated in Fig. 5a and b, respectively.

It is important to note that nanoindentation and tensile tests donot measure the same properties. Generally nanoindentationmodulus (Elastic modulus) values, using the standard O&P method,are higher than results from standard tensile tests. Several reasonshave been invoked to explain this: Surface effects (the initial part ofthe loadedisplacement curve is affected by surface roughness,surface oxidation [38] and other surface phenomena), the type ofloading is compressive in nanoindentation (not tensile), the testfrequencies are quite different (70 Hz for nanoindentation,compared to a much lower frequency in the tensile test) andhydrostatic pressure generated below the Berkovich indenter [39].

Nanoindentation has not been widely used to examine maxil-lofacial polymers, although elastic modulus which is obtained fromnanoindentation is a measure of surface hardness and elasticity aswell as nano-Hardness. These properties also provide significantinformation for maxillofacial materials and for the deteriorationdue to aging, cleaning solutions and skin secretions that firstreflected in the surface and then affect the bulk of the material.Thus, nanoindentation tests are highly important as far as theserviceability of maxillofacial prosthetics is concerned.

The serviceability of prosthesis is one of the major concerns inmaxillofacial surgery. It is very important for patients to be giventhe adequate information concerning the expected averagelongevity of their prostheses, along with information on factorsaffecting the longevity (i.e., environmental staining, cosmetics andcleaning regimes) [40]. Thus, the results from this study could bevery helpful in order for someone to draw conclusions about themechanical behaviour of prosthesis after outdoor weathering. Inthis study, mechanical behaviour of maxillofacial prosthetics,including polydimethylsiloxanes, was affected by irradiation andother climate conditions, after outdoor weathering, as also reportedin other studies [21,41e45]. The alterations which occur in theexamined properties before and after weathering, are in accor-dance with previous studies concerning materials’ structure[21,36,46]. The weathering in different location, if possible is sug-gested due to the variety of weathering characteristics amongdifferent locations [12]. The results show that the examined prop-erties present a similar trend in both locations, although the effectwas more severe for the samples which were placed in Athens.

5. Conclusions

The hypothesis investigated was rejected, since according tostatistical analysis significant changes in mechanical propertieswere observed. In order to correlate the irradiation time with themeasured properties, mathematical models were fitted to theexperimental data. The differences in mathematical models’parameters indicated changes in the measured properties whileirradiation time was increased. Moreover, significant differenceswere observed almost in all the measured properties between thesamples in the two weathering locations.

P.N. Eleni et al. / Polymer Degradation and Stability 96 (2011) 470e476476

The observed differences in the mechanical behaviour, betweenthe control (unirradiated samples) and irradiated samples andbetween the two weathering locations, resulted from photo-degradation and hydrolysis that might occur due to weathering.Material A became harder and according to theoretical approaches,which were analyzed above, its network density increased. Sincecross-linking made samples lose their elasticity and become harderwith an increment in their network density it is possible that this isthe dominant mechanism in material A during irradiation.

The study also provides new information about maxillofacialprosthetics serviceability obtained from nanoindentation tests.

Acknowledgements

The authors wish to thank Dr. Anastasia Poupkou and thelaboratory of Atmospheric Physics of Aristotle University of The-ssaloniki for the technical support in these experiments.

References

[1] Khan Z, Gettleman L, Jacobsen C. Conference report: materials research inmaxillofacial prosthetics. Journal of Dental Research 1992;71:1541e2.

[2] Bellamy KE, Waters MGJ. Designing a prosthesis to simulate the elasticproperties of skin. Bio-Medical Materials & Engineering 2005;15(1):21e7.

[3] Li X, Zhao Y, Li S, Liu X, Wu G, Zhen L, et al. Comparison of mechanicalproperties of Cosmesil M511and A-2186 maxillofacial silicone elastomers.Journal of USeChina Medical Science 2007;4:34e7.

[4] Kiat-amnuay S, Waters PJ, Roberts D, Gettleman L. Adhesive retention ofsilicone and chlorinated polyethylene for maxillofacial prostheses. Journal ofProsthetic Dentistry 2008;99(6):483e8.

[5] Lemon JC, Kiatamnuay S, GettlemanL,Martin JW, ChambersMS. Facial prostheticrehabilitation: preprosthetic surgical techniques and biomaterials. CurrentOpinion in Otolaryngology & Head and Neck Surgery 2005;13(4):255e62.

[6] Rosa DS, Angelini JMG, Agnelli JAM, Mei LHI. The use of optical microscopy tofollow the degradation of isotactic polypropylene (iPP) subjected to naturaland accelerated ageing. Polymer Testing 2005;24(8):1022e6.

[7] Polyzois G, Stafford GD, Winter R. A study of some mechanical properties of anRTV polydimethylsiloxane for extraoral maxillofacial prostheses. ClinicalMaterials 1992;9(1):21e9.

[8] Lacoste-Ferré M, Demont P, Dandurand J, Dantras E, Blandin M, Lacabanne C.Thermo-mechanical analysis of dental silicone polymers. Journal of MaterialsScience 2006;41(22):7611e6.

[9] Dollase T, Spiess HW, Gottlieb M, Yerushalmi-Rozen R. Crystallization ofPDMS: the effect of physical and chemical crosslinks. Europhysics Letters2002;60(3):390e6.

[10] Abbasi F, Mirzadeh H, Katbab A. Modification of polysiloxane polymers forbiomedical applications: a review. Polymer International 2001;50(12):1279e87.

[11] Virlogeux F, Bianchini D, Delor JF, Baba M, Lacoste J. Evaluation of cross-linking after accelerated photo-ageing of silicone rubber. Polymer Interna-tional 2004;53(2):163e8.

[12] Tran NH, Scarbecz M, Gary JJ. In vitro evaluation of color change in maxillo-facial elastomer through the use of an ultraviolet light absorber anda hindered amine light stabilizer. Journal of Prosthetic Dentistry 2004;91(5):483e90.

[13] Gijsman P, Hennekens J, Janssen K. Comparison of UV degradation chemistryin accelerated (xenon) aging tests and outdoor tests (II). Polymer Degradationand Stability 1994;46:63e74.

[14] Dootz ER, Koran A, Craig RG. Physical properties of three maxillofacialmaterials as a function of accelerated aging. Journal of Prosthetic Dentistry1994;71(4):379e83.

[15] Gulmine JV, Janissek PR, Heise HM, Akcelrud L. Degradation profile of poly-ethylene after artificial accelerated weathering. Polymer Degradation andStability 2003;79(3):385e97.

[16] Sampers J. Importance of weathering factors other than UV radiation andtemperature in outdoor exposure. Polymer Degradation and Stability 2002;76(3):455e65.

[17] Pospisil J, Pilar J, Billingham NC, Marek A, Horak Z, Nespurek S. Factorsaffecting accelerated testing of polymer photostability. Polymer Degradationand Stability 2006;91:417e22.

[18] Eleni PN, Katsavou I, Krokida MK, Polyzois GL. Color stability of facial siliconeprosthetic elastomers after artificial weathering. Dental Research Journal2008;5(2):71e9.

[19] Eleni PN, Krokida MK, Frangou MJ, Polyzois GL, Maroulis ZB, Marinos-Kouris D. Structural damages of maxillofacial biopolymers under solar aging.Journal of Materials Science: Materials in Medicine 2007;18(9):1675e81.

[20] Eleni PN, Krokida MK, Polyzois GL. The effect of artificial accelerated weath-ering on the mechanical properties of maxillofacial polymers PDMS and CPE.Biomedical Materials 2009;4(3):035001.

[21] Eleni PN, Katsavou I, Krokida MK, Polyzois GL, Gettleman L. Mechanicalbehaviour of facial prosthetic elastomers after outdoor weathering. DentalMaterials 2009;25(12):1493e502.

[22] Krokida MK, Karathanos VT, Maroulis ZB. Compression analysis of dehydratedagricultural products. Drying Technology 2000;18:395e408.

[23] Krokida MK, Kiranoudis CT, Maroulis ZB. Viscoelastic behaviour of dehydratedproducts during rehydration. Journal of Food Engineering 1999;40(4):269e77.

[24] Oliver WC, Pharr GM. Improved technique for determining hardness andelastic modulus using load and displacement sensing indentation experi-ments. Journal of Materials Research 1992;7(6):1564e80.

[25] Carrillo F, Gupta S, Balooch M, Marshall SJ, Marshall GW, Pruitt L, et al.Nanoindentation of polydimethylsiloxane elastomers: effect of crosslinking,work of adhesion, and fluid environment on elastic modulus. Journal ofMaterials Research 2006;21(2):535e7.

[26] Klapperich C, Komvopoulos K, Pruitt L. Nanomechanical properties of poly-mers determined from nanoindentation experiments. Journal of Tribology2001;123(3):624e31.

[27] Grunlan JC, Xia X, Rowenhorst D, Gerberich WW. Preparation and evaluationof tungsten tips relative to diamond for nanoindentation of soft materials.Review of Scientific Instruments 2001;72(6):2804e10.

[28] Carrillo F, Gupta S, Balooch M, Marshall SJ, Marshall GW, Pruitt L, et al.Nanoindentation of polydimethylsiloxane elastomers: effect of crosslinking,work of adhesion, and fluid environment on elastic modulus. Journal ofMaterials Research 2005;20(10):2820e30.

[29] Meincken M, Berhane TA, Mallon PE. Tracking the hydrophobicity recovery ofPDMS compounds using the adhesive force determined by AFM force distancemeasurements. Polymer 2005;46(1):203e8.

[30] Brogly M, Noel O, Awada H, Castelein G, Schultz J. A nanoscale study of theadhesive contact. Comptes Rendus Chimie 2006;9(1):99e110.

[31] Waters M, Jagger R, Polyzois G, Williams K. Dynamic mechanical thermalanalysis of maxillofacial elastomers. Journal of Prosthetic Dentistry 1997;78(5):501e5.

[32] Rabek JF. Polymer photodegradation: mechanisms and experimental. N. York:Chapman and Hall; 1995. p. 24e5.

[33] Dootz ER, Koran A, Craig RG. Physical property comparison of 11 soft denturelining materials as a function of accelerated aging. Journal of ProstheticDentistry 1993;69(1):114e9.

[34] Guo J-H. Aging processes in pharmaceutical polymers. Pharmaceutical Science& Technology Today 1999;2(12):478e83.

[35] Maxwell RS, Cohenour R, Sung W, Solyom D, Patel M. The effects of [gamma]-radiation on the thermal, mechanical, and segmental dynamics of a silicafilled, room temperature vulcanized polysiloxane rubber. Polymer Degrada-tion and Stability 2003;80:443e50.

[36] Kroschwitz JI, Mark HF. Radiation chemistry of polymers. N.Y.: John Wiley &Sons; 2005.

[37] Flory PJ. Principles of polymer chemistry. Ithaca, NY: Cornell University Press;1953.

[38] Briscoe BJ, Savio Sebastian K. The elastoplastic response of poly(methylmethacrylate) to indentation. Proceedings of the Royal Society A: Mathe-matical, Physical and Engineering Sciences 1996;452(1946):439e57.

[39] Hu Y, Shen L, Yang H, Wang M, Liu T, Liang T, et al. Nanoindentation studies onNylon 11/clay nanocomposites. Polymer Testing 2006;25(4):492e7.

[40] Hooper SM, Westcott TP, Evans LL, Bocca AP, Jagger DC. Implant-supportedfacial prostheses provided by a maxillofacial unit in a U.K. regional hospital:longevity and patient opinions. Journal of Prosthodontics 2005;14(1):32e8.

[41] Feldman D. Polymer weathering: photo-oxidation. Journal of Polymers andthe Environment 2002;10(4):163e73.

[42] White JR, Turnbull A. Weathering of polymers: mechanisms of degradationand stabilization, testing strategies and modelling. Journal of MaterialsScience 1994;29(3):584e613.

[43] Haug SP, Moore BK, Andres CJ. Color stability and colorant effect on maxil-lofacial elastomers. Part II: weathering effect on physical properties. Journal ofProsthetic Dentistry 1999;81(4):423e30.

[44] Vallis E, Sarmoria C, Villar M, Lazzari M, Chiantore O. Model poly-dimethylsiloxanes subjected to thermal weathering: effect on molecularweight distributions. Polymer Degradation and Stability 2000;69(1):67e71.

[45] Bellamy K, Limbert G, Waters MG, Middleton J. An elastomeric material forfacial prostheses: synthesis, experimental and numerical testing aspects.Biomaterials 2003;24:5061e6.

[46] Shah H, Munisamy S, Vaidyanathan TK. Artificial aging of chlorinated poly-ethylene and maxillofacial silicone elastomer. In: IADR 86th general session &exhibition, Toronto; 2008.