Polypropylene_cycloolefin copolymer blends_ effect

structure on tensile mechanical properties

Alessandro Pegoretti a,*,Jan Koları

´k b ,Luca Fambri a ,Amabile Penati a a

Department of Materials Engineering and Industrial Technologies,University of Trento,Via Mesiano 77,38050Trento,Italy

b

Institute of Macromolecular Chemistry,Academy of Sciences of the Czech Republic,16206Prague 6,Czech Republic

Received 26September 2002;received in revised form 13February 2003;accepted 11March 2003

Abstract

Tensile mechanical properties of polypropylene (PP)/cycloolefin copolymer (COC)blends were studied using an Instron tensile tester.As COC was expected to impart enhanced mechanical properties to the blends,their modulus,yield strength,tensile strength and tensile energy to break were measured as functions of blend composition.With regard to the reported sensitivity of the COC structure to thermal history,the influence of annealing at two different temperatures was also tested.The attention was primarily concentrated on blends with the volume fraction of COC in the interval 0,v 2,0:40;where COC formed (short)fibres almost uniaxially oriented in the direction of injection moulding.In the interval 0:40,v 2,0:75;the blends consisted of partially co-continuous components.Two different models were applied in the analysis of mechanical properties,namely (i)the rule of mixtures for fibre composites and (ii)the equivalent box model for isotropic blends (employing the data on the phase continuity of components obtained from modified equations of the percolation theory).Experimental data on the studied mechanical properties were better fitted by the models for fibre composites.Annealing of the samples (758C for 45days;1208C for 3h)did not markedly affect the tensile modulus,yield stress,and stress at break of the blends.On the other hand,the strain at break was markedly reduced by the annealing up to v 2¼0:2;COC and the blend with 75%of COC ruptured in a brittle manner without yielding.q 2003Elsevier Science Ltd.All rights reserved.

Keywords:Polymer blends;Polypropylene;Cycloolefin copolymer

1.Introduction

Preparation of polymer blends is one of the most cost-effective ways for the upgrading of existing polymers [1–6].As potential applications of various polymeric materials are codetermined by their mechanical properties,it is desirable to know the relationships between the morphology (phase structure)and physical properties of intended blends.As generally known,polypropylene (PP)shows relatively low modulus,yield strength and resistance to creeping.Thus,search for ‘reinforcing’polymeric components,which could be easily blended with PP,is still an interesting problem to be solved.Preparation of blends without compatibilisers may be difficult because the compatibility of PP with other polymers is limited [2,5].Recently,amorphous ethylene–norbornene copolymers obtained with metallocene-based catalysts [7,8]have been marketed [9].They rank among new polymer materials with remarkable properties,such as

a high glass transition temperature ðT g Þ;transparency,heat resistance,chemical resistance to common solvents,low moisture uptake,high water barrier,good mechanical properties,etc.Available products—usually denoted as cycloolefin copolymers (COC)—have recently attracted much attention in the field of basic and applied material science [9–16].Studies of mechanical properties of ethylene–norbornene copolymers encompass dynamic mechanical thermal analysis (DMTA)[13,15],stress–strain measurements [9,13]flexural creep [9],micro-hardness [13],impact strength [10],etc.Rising percentage of norbornene accounts for increase in the yield or tensile strength and decrease in the strain at yielding and break [9,12];the phenomenon of yielding is preserved up to about 40%of norbornene in copolymers.Interestingly enough,increase in T g caused by annealing was attributed [12]to the growth of rigid amorphous phase due to the short-range ordering of norbornene chain segments.

Because of its olefinic character,COC is expected [17,18]to be compatible with polypropylene and other polyolefins;for this reason,we have attempted to prepare [16]the PP/COC blends without special compatibilisers.Of

0032-3861/03/$-see front matter q 2003Elsevier Science Ltd.All rights reserved.

doi:10.1016/S0032-3861(03)00248-9

Polymer 44(2003)3381–3387

www.elsevier.com/locate/polymer

*Corresponding author.Tel.:þ39-0461-882452;fax:þ39-0461-881977.

E-mail address:alessandro.pegoretti@ing.unitn.it (A.Pegoretti).

In our previous paper[16],phase morphology of PP/ COC blends was studied by means of the scanning electron microscopy(SEM)and scanning transmission electron microscopy(STEM).Looking for the critical volume fraction of COC,at which this component assumes(partial) continuity,our interest was mainly focused on the composition interval up to50%COC.Surprisingly enough, studied blends were found to havefibrous morphology.In the90/10,80/20and70/30blends,the PP matrix contained fibres of COC,whose average diameter increased with the COC fraction in the range0.25–0.80m m.In the60/40 blend,the COC component formed bothfibres(average diameter2.6m m)and larger elongated entities in the PP matrix.The50/50blend consisted of co-continuous COC and PP components,while in the25/75blend,PPfibres were embedded in COC matrix.In all blends,thefibres were almost uniaxially oriented in the injection direction.Many COCfibres were partly pulled out from the PP matrix on the fracture surfaces perpendicular to the direction of injection. Some COCfibres were broken at the level of fractured surface,which evidences a noticeable interfacial adhesion between PP and COC.The microphotographs thus indicated that COCfibres were long enough to be broken instead of pulled out from the PP matrix at existing interface adhesion. According to the available literature,spontaneously formed and stablefibrous morphology of polymer blends is rather exceptional.COCfibres were not a product of additional drawing,but they were formed during mixing and/or injection moulding(the latter process brought about the uniaxial orientation of the COCfibres).An averagefibre aspect ratio was estimated to be at least20.

As COC forms shortfibres in our PP/COC blends[16, 36],we consider measurement and analysis of the tensile mechanical properties of‘fibrous’blends a new and interesting topic.Such blends are expected to have proper-ties different from those of the‘standard’isotropic heterogeneous blends.The objective of this paper is(i)to estimate the effects of thefibrous phase structure on tensile properties,(ii)to compare these properties,wherever possible,with the prediction of existing models,and(iii) to check the influence of annealing on the tensile mechanical properties of the prepared blends.

2.Models used

2.1.Equivalent box model for heterogeneous isotropic materials

Standard polymer blends are heterogeneous isotropic materials with the three-dimensional continuity of one or more components so that simple parallel or series models or the models for orthotropic or quasi-isotropic materials are not applicable.In our previous papers[19–21,27,28,37,38] we have proposed and verified a predictive format based on the combination of the equivalent box model(EBM)and the concept of phase continuity.The EBM in Fig.1operates with partly parallel(subscript p)and partly serial(subscript s)couplings of components.This EBM is a two-parameter model as of four volume fractions v ij only two are independent;its volume fractions are interrelated as follows:

v1¼v1pþv1s;v2¼v2pþv2s;

v1þv2¼v pþv s¼1

ð1Þ

The blocks in the EBM are presumed to have physical properties of the neat components;the EBM is likely to fail if the mixing process produces a significant change in the structure and properties of a constituent.As the EBM is

not

Fig.1.Equivalent box model for a binary blend(schematically).

A.Pegoretti et al./Polymer44(2003)3381–3387 3382a self-consistent model,the predictive format requires two

steps:(1)to derive the equations for the properties under

consideration;(2)to calculate v ij by using another appro-

priate model,e.g.modified equations[19–21,27,28,37,38]

rendered by the percolation theory[39–41].An essential

feature of the proposed scheme is that all simultaneously

predicted properties of a blend are related to a certain phase

structure through an identical set of input parameters.

Tensile moduli of the parallel and series branches of the

EBM[19–21,28]are the following:E p¼ðE1v1pþ

E2v2pÞ=v p;E s¼v s=½ðv1s=E1Þþðv2s=E2Þ :The resulting ten-sile modulus of two-component systems is then given as the

sumðE p v pþE s v sÞ:

E b¼E1v1pþE2v2pþv2s=½ðv1s=E1Þþðv2s=E2Þ ð2ÞA linear stress–strain relationship indispensable for the modulus measurements can be granted for glassy and/or crystalline polymers only at very low strains,typically below1%,where virtually all blends show interfacial adhesion sufficient for the transmission of acting stress.At higher strains(usually3–6%),the applied tensile stress exceeds the linearity limit and attains the value of yield strength,thus inducing yielding and plastic deformation of constituents.In our previous papers[19–21,28],we have derived the following equation for S yb of the EBM visualised in Fig.1:

S yb¼S y1v1pþS y2v2pþAS y1v sð3Þwhere S y1,S y2characterise the parent polymers and A the extent of interfacial debonding.Two limiting values of S yb; identified with the lower or upper bound,can be distinguished by means of Eq.(3):(i)interfacial adhesion is so weak that a complete debonding occurs between the fractions of constituents coupled in series before yielding is initiated(A¼0at the yield stress);(ii)interfacial adhesion is strong enough to transmit the acting stress between constituents so that no debonding appears in the course of yieldingðA¼1Þ:However,if two components differing in the yield strength are coupled in series,the series branch (Fig.1)yields at S y1or S y2;whichever is lower. Experimental experience[28,42,43]shows that formally analogous equations can be used for evaluation of the yield as well as tensile strength of particulate systems;thus,we will tentatively apply Eq.(3)for S ub by replacing S y1and S y2 by the tensile strengths S u1and S u2;respectively.

Employing a universal formula provided by the percola-

tion theory[39]for the elastic modulus of binary systems,

we can calculate[19–21,28]v ij by using the following

equations:

v1p¼½ðv12v1crÞ=ð12v1crÞ qð4aÞv2p¼½ðv22v2crÞ=ð12v2crÞ qð4bÞwhere v1cr or v2cr is the critical volume fraction(the percolation threshold)at which the component1or2 becomes partially continuous and q is the critical exponent;the remaining v1s and v2s are evaluated by using Eq.(1).In the marginal zone0,v1,v1cr(or0,v2,v2cr),where only component2(or1)is continuous,simplified relations can be used for the minority component,i.e.v1p¼0;v1s¼v1(or v2p¼0;v2s¼v2),to obtain an approximate prediction of mechanical properties.Most ascertained values of q are located in an interval of1.6–2.0so that q¼1:8can be used also as an average value[39–41].For three-dimensional cubic lattice,the percolation threshold v cr¼0:156was calculated[40,41].In general,the patterns predicted by using‘universal’values v1cr¼v2cr¼0:156 and q1¼q2¼1:8should be viewed as afirst approxi-mation that may not be in a good accord with experimental data because real v1cr and v2cr of polymer blends frequently differ from0.156and from each other.

2.2.Models forfibrous structures

As we have reported in our previous paper[16],prepared PP/COC blends with weight fractions of COC up to0.4have a structure resembling shortfibre composites.Thus in this interval the models for(short)fibre composites are to be applied.Longitudinal tensile modulus of aligned shortfibres is routinely calculated by means of the Halpin–Tsai equation[42,44–46]that is usually expressed in the following form:

E b¼E1ð1þABv2Þ=ð12Bv2Þð5aÞwhere A¼2L=d is given by the ratio of thefibre length L and thefibre diameter d;while B¼½ðE2=E1Þ2

1 =½ðE2=E1ÞþA :For very high aspect ratios,Eq.(5a)is reduced[45]to the rule of mixtures

E b¼E1v1þE2v2ð5bÞSimilarly enough,the tensile strength S ub of a composite with uniaxially oriented shortfibres reads[45]

S ub¼S01v1þS u2v2½12ðL cr=2LÞ ð6aÞwhere S u2is the tensile strength offibres,S01is the stress attained in the matrix when the strain-at-break offibres is reached,and L cr is the critical length of shortfibres. Obviously,if L@L cr;Eq.(6a)can be simplified as follows (rule of mixture):

S ub¼S01v1þS u2v2ð6bÞThe minimum volume fraction that ensuresfibre-controlled composite failure is usually indicated as v2min[44,45]:

v2min¼ðS1u2S01Þ=½ð12L cr=LÞS2uþS1u2S01 ð7aÞwhere S1u is the tensile strength of the matrix.For this volume concentration of shortfibres,S ub passes through a minimum.

For L@L cr;Eq.(7a)is reduced to the form valid for the composites with continuousfibres:

v2min¼ðS1u2S01Þ=ðS2uþS1u2S01Þð7bÞ

A.Pegoretti et al./Polymer44(2003)3381–33873383As long as v2,v2min;the composite will not fracture when allfibres break because the remaining matrix cross-section can still support the load.As a consequence the composite strength will be lower than the matrix strength and given by:

S ub¼S1uð12v2Þ¼S1u v1ð7cÞ

For v2.v2min;the composite strength can be predicted by using Eq.(6b).

3.Experimental

3.1.Materials

Polypropylene Moplen C30G was the product of Basell, Ferrara,Italy:meltflow index(2308C, 2.16kg)¼5.7 g/10min;density:0.903g/cm3;crystallinity:52%; T g¼2108C.Amorphous cycloolefin copolymer produced under the trade name Topas8007was the product of Ticona-Celanese,Frankfurt,Germany,consisting of30%of norbornene and70%of ethylene:MFI(1908C,

2.16kg)¼1.7g/10min;density:1.020g/cm3;T g¼758C.

3.2.Blend preparation

A series of the PP/COC blends was prepared with5,10, 15,20,25,30,40,50,75weight%of COC.Polymers were mixed in a Banbury mixer(chamber4.3l;1rpm)at 1908C for3.5min.Produced pellets were used for feeding a Negri–Bossi injection moulding machine to produce test specimens for the measurements of mechanical properties. Two types of the test pieces were prepared:(1)ASTM D638 (length:210mm;thickness:3.3mm;gauge length:80mm; gauge width:12.8mm;barrel temperature:2428C;injec-tion pressure:20MPa);(2)ISO527(170;4;80;10mm; 2158C;30MPa).Specimens used for the testing of mechanical properties were stored for more than6months at room temperature to avoid any interfering effect of the physical ageing during measurements.3.3.Stress–strain measurements

Instron tensile tester,model4502,was used to measure tensile mechanical properties of studied blends.Tensile modulus was determined by using a strain gauge extens-ometer(Instron,model2620;gauge length:25mm)on ASTM dog-bone shape specimens tested up to1%strain at the cross-head speed of1mm/min(three specimens were tested for each blend).Tensile yield strength and strain, stress and strain at break and tensile energy to break were also ascertained on the ASTM test pieces tested up to the fracture at the cross-head speed of40mm/min,i.e.at the strain rate of50%/min(six specimens were measured for each blend).All test were carried out at about258C.

3.4.Annealing of PP/COC blends

Two different types of thermal treatment were used:(1) storage at758C(i.e.at about T g of COC)for45days followed by cooling at a rate of28C/h;(2)storage at1208C (i.e.about458C above T g of COC)for3h followed by a fast cooling to room temperature(at a cooling rate of

about Fig.2.Stress–strain curves of(a)PP,(b)PP/COC¼50/50and(c)

COC.

Fig.3.Tensile modulus as a function of the COC fraction in blends.Full

line:the rule of mixtures(5b);dashed line:the EBM(2)for v1cr¼0:16;

v2cr¼0:103and q¼1:2;dotted line:q¼1:8

:

Fig.4.Yield stress as a function of the COC fraction in blends.Full line:the

rule of mixtures;dashed line:the EBM(3)for A¼1;v1cr¼0:16;v2cr¼

0:103and q¼1:2;dotted line:q¼1:8:

A.Pegoretti et al./Polymer44(2003)3381–3387

3384

28C/s).The treatment (1)was intended to imitate accelerated physical ageing in order to obtain preliminary information on the material properties after a long storage of blends at ambient temperatures.The second type of treatment was conducted to verify the resistance of blends to short-term action of elevated temperatures (temperature of annealing 1208C was still much lower than the melting temperature of PP).It is important to note that the shape and dimensions of the test specimens were not changed by any of these treatments.

4.Results and discussion

Typical stress–strain curves found for PP,COC and their blend 50/50are given in Fig.2.PP has modulus E 1¼1:47^0:03GPa,yield stress S y1¼32:2^0:3MPa,yield strain e y1¼11:5^0:2%;stress-at-break S u1¼25:5^2:9MPa and strain-at-break e u1¼684^87%:On the other hand,COC as a ‘reinforcing’component shows E 2¼2:82^0:04GPa,S y2¼63:9^0:7MPa e y2¼4:8^0:3%;S u2¼35:3^1:4MPa and e u2¼10^2%:The 50/50blend is characterised by E b ¼2:21^0:16GPa,S yb ¼45:3^1:2MPa,e yb ¼4:8^0:2%;S ub ¼28:9^0:9MPa and e ub ¼27^6%;which

clearly evidences the contribution of the COC component to the tensile properties of the resulting blend.

Tensile modulus E b as a function of blend composition (Fig.3)can plausibly be fitted by the rule of mixtures (5b)valid for the composites with uniaxially oriented continuous fibres,while the prediction (2)of the EBM is lower.Thus the COC fibres spontaneously created in blends (for v 2,0:5)seem to be long enough to impart to the blends enhanced stiffness similar to that of long-fibre composites.The co-continuous structures constituted at 0:5#v 2#0:75have the moduli also very close to those predicted by the rule of mixtures,which can be explained by the observation [16]that,depending on the composition,the co-continuous PP component contains COC fibres,while the co-continu-ous COC component contains PP fibres.It is to be noted that a relatively small difference between the tensile moduli of components accounts for the fact that the modulus E b predicted by the EBM is only slightly lower than that predicted by the rule of mixtures.

Yield stress S yb predicted by Eq.(3)for the EBM is expected to start increasing at v 2$v 2cr ;when the second components with S y2.S y1becomes co-continuous.The experimental data in Fig.4are fitted by Eq.(3)quite

Fig.5.Stress at break as a function of the COC fraction in blends.(a)The EBM (3)for v 1cr ¼0:16;v 2cr ¼0:103and q ¼1:2:full line:A ¼1;dashed line:A ¼0:5;dotted line:A ¼0:(b)Full lines:the rule of mixtures

(7)

Fig.6.Yield strain (full points)and strain at break (empty points)as functions of the COC fraction in

blends.

Fig.7.Strain at break as a function of the COC fraction in aged blends.Full points:as-received specimens;empty points:annealing at 758C for 45days,empty triangles:annealing at 1208C for 3h.

A.Pegoretti et al./Polymer 44(2003)3381–33873385

Tensile strength S ub of the PP/COC blends(Fig.5a) passes through a minimum at v2¼0:136ðw2¼0:15Þ; which cannot befitted by the EBM(Eq.(3)).However, if only the blends with thefibres of COC are considered (Fig.5b),i.e.v2,0:5;the experimental data obey quite well Eqs.(6b)and(7c)forfibre composites.Unfortu-nately,a quantitative analysis of the data is difficult because the values of S2u and e2u for the COCfibres spontaneously formed in the course of blend preparation are not known.An extrapolation of S2u in Fig.5b leads to a value of about39MPa,which is higher than the value of35.3MPa found for COC in bulk.This result is reasonable considering thefibrous structure of COC in the blends up to v2¼0:5:

Yield strain of the blends markedly decreases with the fraction of COC(Fig.6)in the region v2,0:3and remains virtually constant for v2.0:4;where it is equal to the value for neat COC.On the other hand,strain at break remains very high(almost equal to that of PP)for v2¼0:1; afterward,it drops rapidly to the value characterising COC.Thus,both yield strain and strain at break concur-rently reflect the changes in the phase structure of the blends.

4.1.Effects of annealing on tensile properties of PP/COC blends

Table1summarises the values of the tensile modulus, yield stress and stress at break obtained from the tensile tests conducted on the samples annealed at758C for45days and at1208C for3h.The selected annealing conditions do not have a significant systematic effect on these mechanical properties(for comparison see Figs.3–6).Moreover it is interesting to observe that the trend of the stress at break as a function of the COC content still passes through a minimum,which suggests that the annealing treatments do not modify thefibrous structure of COC in the PP matrix. On the other hand,the strain at break was noticeably reduced for the blends in the region v2,0:2;as evidenced by Fig.7.Embrittlement of the annealed blends can be viewed as a result of the reduction of molecular mobility, which is in conformity with previous results[12,47].It is worth noting that the annealing at658C for30days (imitating an ageing treatment)performed with some selected compositions of the PP/COC blends has not induced any appreciable variation of the mechanical properties,including the strain at break.

5.Conclusions

PP and COC were found to be compatible polymers forming—under fortuitously selected conditions of mix-ing—blends withfibrous structures.Although no compati-biliser was used,the adhesion between components was fairly good.Tensile mechanical properties of the PP/COC blends were found to be markedly affected by the blend composition:increasing fraction of‘reinforcing’COC component in the blends accounted for an increase in the modulus,yield strength and tensile strength,while the yield strain,strain at break and tensile energy to break showed a rather intensive drop.

As in the composition interval0,w2#0:40COC formedfibres(almost uniaxially oriented in the direction of injection moulding),while in the interval0:40,w2#0:75 the blends consisted of partially co-continuous components, two different models were applied in the analysis of mechanical properties:(i)the rule of mixtures forfibre composites;(ii)the equivalent box model(EBM)for isotropic blends where the data on the phase continuity of components were obtained from modified equations of the percolation theory.Tensile modulus as a function of blend composition was plausiblyfitted by the rule of mixtures. Monotonic rise of yield strength with the COC fraction was in conformity with the prediction of the EBM under the assumption of good interfacial adhesion.On the other hand, tensile strength passing through a minimum was satisfac-torily described by the model forfibre composites.

With regard to the reported sensitivity of COC to thermal history,the influence of annealing at two different temperatures was also tested.Annealing of samples(758C for45days;1208C for3h)did not profoundly affect the modulus,yield stress and stress at break of blends.On the other hand,the strain at break was markedly reduced; moreover,COC and the blend with75%of COC ruptured in a brittle manner.Thus the annealing accounted for some embrittlement of the blends,probably owing to the reduction of the free volume and molecular mobility. Acknowledgements

The author Jan Koları´k is greatly indebted to the Grant Agency of the Academy of Sciences of the Czech Republic forfinancial support of this work(Grant No.A4050105). The authors are grateful to Dr P.Goberti(Basell,Ferrara, Italy)for supplying polypropylene and for the preparation of blends and test specimens and to Diana Bortot for extensive technical assistance.

A.Pegoretti et al./Polymer44(2003)3381–3387 3386

References

[1]Bucknall CB.Toughened plastics.London:Applied Science Publi-

cation;1977.

[2]Utracki LA.Polymer alloys and blends.Munich:Hanser Publication;

1990.

[3]Folkes MJ,Hope PS.Polymer blends and alloys.Cambridge:

Chapman and Hall;1993.

[4]Sperling LH.Polymeric multicomponent materials.New York:

Wiley;1997.

[5]Paul DR,Newman S.Polymer blends.New York:Academic Press;

1978.

[6]Paul DR,Bucknall CB,editors.Polymer blends.New York:Wiley;

1999.

[7]Tritto I,Boggioni L,Jansen JC,Thorshaug K,Sacchi MC,Ferro DR.

Macromolecules2002;35:616–23.

[8]Thorshaug K,Mendichi R,Boggioni L,Tritto I,Trinkle S,Friedrich

C,Muelhaupt R.Macromolecules2002;35:2903–11.

[9]Topas(COC),brochure from Ticona,Hoechst,Germany,1995.

[10]Khanarian G.Polym Engng Sci2000;40:2590–601.

[11]Huang WJ,Chang FC,Chu PPJ.Polymer2000;41:6095–101.

[12]Chu PPJ,Huang WJ,Chang FC.Polymer2000;42:2185–91.

[13]Scrivani T,Benavente R,Perez E,Perena JM.Macromol Chem Phys

2001;202:2547–53.

[14]Forsyth J,Perena JM,Benavente R,Perez E,Tritto I,Boggioni L,

Brintzinnger HH.Macromol Chem Phys2001;202:614–20.

[15]Forsyth JF,Scrivani T,Benavente R,Marestin C,Perena JM.J Appl

Polym Sci2001;82:2159–65.

[16]Slouf M,Koları´k J,Fambri L,J Appl Polym Sci,submitted for

publication.

[17]Rana D,Lee CH,Cho K,Lee BH,Choe S.J Appl Polym Sci1998;69:

2441–50.

[18]Luettmer-Strathmann J,Lipson JEG.Macromolecules1999;32:

1093–102.

[19]Koları´k J.Polym Engng Sci1996;36:2518–24.

[20]Hora´k Z,Koları´k J,Sı´pek M,Hynek V,Vecerka F.J Appl Polym Sci

1998;69:2615–23.

[21]Koları´k J.Eur Polym J1998;34:585–90.

[22]Matsuoka T,Yamamoto S.J Appl Polym Sci1998;68:807–13.[23]Willemse RC,de Boer AP,van Dam J,Gotsis AD.Polymer1998;39:

5879–87.

[24]Willemse RC.Polymer1999;40:2175–8.

[25]Veenstra H,de Boer J,de Boer AP.Polymer1999;40:1119–30.

[26]Veenstra H,Verkooijen PCJ,van Lent BJJ,Van Dam J,de Boer AP,

Nijhof HJ.Polymer2000;41:1817–26.

[27]Koları´k J,Pegoretti A,Fambri L,Penati A.J Polym Res2000;7:7–14.

[28]Koları´k J,Fambri L,Pegoretti A,Penati A.Polym Engng Sci2000;40:

127–31.

[29]Verhoogt H,Langelaan HC,Van Dam J,De Boer AP.Polym Engng

Sci1993;33:754–63.

[30]Machiels AGC,Denys KFJ,Van Dam J,De Boer AP.Polym Engng

Sci1996;36:2451–66.

[31]Machiels AGC,Denys KFJ,Van Dam J,De Boer AP.Polym Engng

Sci1997;37:59–72.

[32]Boyaud MF,Ait-Kadi A,Bousmina M,Michael A,Cassagnau P.

Polymer2001;42:6515–26.

[33]Evstatiev M,Schultz JM,Fakirov S,Friedrich K.Polym Engng Sci

2001;41:192–204.

[34]Boyaud MF,Cassagnau P,Michel A.Polym Engng Sci2001;41:

684–95.

[35]Zhang L,Huang R,Li L,Wang G.J Appl Polym Sci2002;83:1870–4.

[36]Koları´k J,Pegoretti A,Fambri L,Penati A,submitted.

[37]Koları´k J.J Polym Sci B Polym Phys,J Polym Sci B Polym Phys

2003;41.

[38]Koları´k J,Pegoretti A,Fambri L,Penati A.J Appl Polym Sci2003;88:

1–51.

[39]De Gennes PG.J Phys Lett(Paris)1976;37:L1–L5.

[40]Hsu WY,Wu S.Polym Engng Sci1993;33:293–302.

[41]Sax J,Ottino JM.Polym Engng Sci1983;23:165–76.

[42]Nielsen LE,Landel RF.Mechanical properties of polymers and

composites.New York:Marcel Dekker;1994.

[43]Nicolais L,Narkis M.Polym Engng Sci1971;11:194–9.

[44]Ashton JE,Halpin JC,Petit PH.Primer on composite materials

analysis.Stamford:Technomic;1969.

[45]Agarwal BD,Broutman LJ.Analysis and performance offiber

composites.New York:Wiley;1990.

[46]Halpin JC,Kardos JL.Polym Engng Sci1976;16:344–52.

[47]Luda MP,Ragosta G,Musto P,Pollicino A,Camino G,Recca A,

Nepote V.Macromol Mater Engng2002;287(6):404–11.

Table1

Tensile mechanical properties of various PP/COC blends annealed under different conditions

PP/COC(weight ratio)Annealing conditions

Modulus(MPa)Yield stress(MPa)Stress at break(MPa)

758C/45days1208C/3h758C/45days1208C/3h758C/45days1208C/3h 100/0 1.57^0.05 1.61^0.0332.2^0.333.4^0.122.1^0.223.1^0.1 95/5 1.62^0.05 1.60^0.0934.0^0.634.4^0.120.6^0.422.8^2.0 90/10 1.^0.05 1.73^0.0733.4^0.634.4^0.419.2^0.919.8^0.1 85/15 1.70^0.05 1.70^0.0434.0^1.035.2^0.120.8^0.320.8^0.1 80/20 1.77^0.04 1.67^0.0134.8^0.336.3^0.222.3^0.222.8^0.1 75/25 1.83^0.09 1.77^0.0836.2^0.137.5^0.423.4^1.524.5^0.1 70/30 1.86^0.03 1.^0.1938.1^0.338.8^0.125.1^0.124.5^0.4 60/40 1.94^0.03 1.84^0.0240.5^0.141.4^0.526.8^0.726.3^0.2 50/50 2.06^0.05 1.96^0.0245.4^0.145.9^0.428.9^0.428.0^0.2 75/25 2.53^0.01 2.35^0.0955.3^0.2n.m.57.1^0.555.0^1.0 0/100 2.60^0.07n.a..0^0.7n.a.61.7^1.0n.a.

n.m.¼not measurable;n.a.¼not available(the annealing temperature higher than the glass transition temperature of COC accounted for warping of test specimen.

A.Pegoretti et al./Polymer44(2003)3381–33873387