Rheological properties of ground tyre rubber based thermoplastic elastomeric blends

Polymer Testing 45 (2015) 58e67

Contents lists available at ScienceDirect

Polymer Testing journal homepage: www.elsevier.com/locate/polytest

Material properties

Rheological properties of ground tyre rubber based thermoplastic elastomeric blends ~es da Silva a, Jose  Oliveira a, b, Vítor Costa c Paulo Lima a, *, Sara P. Magalha a

School of Design, Management and Production Technologies, University of Aveiro, Estrada do Cercal, 449, 3720-509 Santiago de Riba-Ul, Portugal rio de Santiago, 3810-193 Aveiro, Portugal CICECO, University of Aveiro, Campus Universita c rio de Santiago, 3810-193 Aveiro, Portugal Mechanical Engineering Department, University of Aveiro, Campus Universita b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 15 April 2015 Accepted 21 May 2015 Available online 28 May 2015

This work analyses the rheological behaviour of thermoplastic elastomeric blends (TPE) based on ground tyre rubber (GTR), more specifically the rheological behaviour of binary and ternary polypropylene (PP) based blends with different rubber materials: an ethylene propylene diene monomer (EPDM), an ethylene propylene rubber (EPR) and GTR. The study was developed under steady-shear rate conditions by capillary rheometry at three different temperatures. TimeeTemperature Superposition Principle (TTSP) was applied to the viscosity curves using a temperature dependent shift factor, allowing the construction of master curves for the analysed blends. The Cross-WLF model was used to predict the rheological parameters, giving numerical results for viscosity similar to the experimental data. GTR increased the blends viscosity. EPR showed rheological behaviour similar to PP, and EPDM presented higher power law behaviour. Pseudoplastic behaviour was observed for all the analysed blends. Incorporation of GTR in TPE blends for injection moulding purposes was found to be a feasible strategy to upcycle this type of potentially wasted material. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Thermoplastic elastomeric blends Ground tyre rubber Recycling Rheological properties Cross-WLF model

1. Introduction Sustainable waste management policies have been implemented in recent decades to deal with end of life tyres (ELT). The use of complete, shredded or ground tyres in civil engineering projects, sports fields and moulded products are some of the implemented strategies to reutilise this potentially wasted product [1,2]. The use of injection moulding to recycle GTR, a by-product of ELT, can be a viable approach to upcycle this potential waste. For this purpose, the performance of thermoplastic blends with GTR has been studied by several authors [3]. However, due to the GTR being vulcanized and its lack of compatibility with polyolefins, the mechanical performance of these blends is still a limiting factor. Several strategies have been employed to improve the compatibility, such as regeneration techniques and the use of compatibilizing agents. TPE blends based on GTR, fresh rubber and polyolefin matrix (TPEGTR) have also been studied, with special emphasis on their compatibility and morphology [4e13]. The majority of the

* Corresponding author. E-mail address: [email protected] (P. Lima). http://dx.doi.org/10.1016/j.polymertesting.2015.05.006 0142-9418/© 2015 Elsevier Ltd. All rights reserved.

studies has been focused on the mechanical properties of the blends and very few on their processability, which is a fundamental aspect for the development of TPEGTR blends for the injection moulding industry. Nowadays, numerical simulation of the injection moulding process is a powerful tool for the production of high quality products with enhanced productivity. The application of such simulation tools on the development of moulds for the injection of TPEGTR blends requires the characterisation of these blends as regards flow and heat transfer. Rheology of polymer blends plays one of the most significant roles in the reliability of the numerical simulations and it is strongly influenced by the nature of the materials, their compatibility as well as their viscosity. A rheological study is, therefore, crucial to understand the influence of the different materials and compositions on the flowability of these blends under different processing conditions. By providing useful information about the material interactions and blend morphology, it can also contribute to better comprehension of their mechanical behaviour. Prut et al. [14] analysed the dynamic rheological properties of binary PP/GTR composites containing different weight contents of GTR and PP with different molecular weight characteristics. Pseudoplastic behaviour was detected for all the blends with increasing

P. Lima et al. / Polymer Testing 45 (2015) 58e67

GTR content, and a strong deviation from the Newtonian regime. Costa et al. [15] analysed the rheological behaviour of ternary blends based on low density polyethylene (LDPE) with EPDM and GTR in dynamic mode. They observed a decrease of viscosity with increasing frequency, characteristic of pseudoplastic behaviour, although they did not detect a Newtonian plateau even at low frequencies. GTR did not led to substantial increase of viscosity, which was explained by a probable encapsulation of GTR by EPDM. Kim et al. [16] studied the effects of GTR ultrasonic treatment and a compatibilizing agent on the rheological behaviour of polyolefin/ GTR blends. The blends indicated pseudoplastic behaviour and the viscosity revealed higher dependence on the polyolefin material rather than on the GTR treatment and compatibilizing agent. The influence of different compatibilizers and bitumen on PP/GTR blends was analysed by Zhang et al. [17]. Bitumen was found to have a plasticizing effect, whereas the compatibilizers increased the viscosity. They concluded also that the polar compatibilizers have a higher effect on the viscosity increase than the non-polar compatibilizers. Scafaro et al. [18] analysed the processability of polyolefin/GTR blends prepared in a twin screw extruder without any additives. Different extrusion parameters such as temperature and mixing speed were set trying to promote the GTR thermomechanical devulcanization and, therefore, better compatibilization and mechanical properties. Results indicated a viscosity increase and most pronounced non-Newtonian behaviour for increasing GTR content. Different extrusion parameters and mixing procedures, such as mixing speed and mixing steps have only small effects on the blend rheology. A high processing temperature (300
C) was found to have some disrupting effect on the threedimensional network of the crosslinked GTR, and could also lead to some thermal degradation. This work is part of an ongoing study that aims to contribute on the sustainability of the GTR recycling process, through the development of GTR based blends for injection moulding, without resorting to thermochemical methods. The mechanical and thermal properties of TPEGTR blends based on a high melt flow PP, suitable for injection moulding of thin or/and complex parts, have already been studied by the same authors [19,20]. EPDM and EPR were chosen as the fresh rubber materials due to their toughening effect on the PP based blends and to their compatibility with GTR. The purpose of the present work is to study the processability of TPEGTR blends by evaluating their rheological behaviour. The effects of the blend composition is analysed by capillary rheometry. The Cross-WLF model is used to predict the flow behaviour of these TPEGTR blends. 2. Rheological analysis

Capillary rheology allows determination of the material apparent viscosity (ha ) defined as

ha ¼

ta ðPa$sÞ g_ a

(1)

where ta is the apparent wall shear stress and g_ a the apparent wall shear rate. These parameters can be calculated accordingly to the Poiseuille Law, based on the barrel diameter, plunger speed and length and diameter of the capillary. The apparent wall shear rate is given by

g_ a ¼

4Q
1  s pR3

where Q is the volumetric flow rate of the polymer melt and R the capillary die radius. The volumetric flow rate can be calculated as


 Q ¼ pR2b Sp mm3 :s1

(2)

(3)

where Rb (mm) is the radius of the barrel and Sp (mm.s1) is the plunger speed. The pressure drop, DP ðPa), measured across the capillary length, L (mm) is used to determine the apparent shear stress as

ta ¼ DP

R ðPaÞ 2L

(4)

This model assumes fully developed flow along the entire capillary length, disregarding the extra pressure drop at the entrance of the capillary die. The entrance and exit effects on the rheological data can be corrected using Bagley's correction [21] which allows the determination of the true wall shear stress (tw )

tw ¼

DP  DPe ðPaÞ 2 L=R

(5)

DPe e pressure drop at zero distance from the entrance. The shear rate expression assumes a Newtonian parabolic velocity profile. Due to the polymer non-Newtonian behaviour, the real profile is non-parabolic, similar to a plug-like flow. Assuming no slip conditions, the velocity is higher at the centerline and zero at the wall, which implies the highest shear rate at the wall. The profile shape is defined by a power law index (n) which characterizes the pseudoplastic behaviour of the material. n values below 1 represent the transition from Newtonian flow to shear thinning behaviour. Smaller values imply higher shear thinning behaviour, thus a greater deviation from the (Newtonian) parabolic profile [22]. The Weissenberg-Rabinowitsch correction [23] can be used to determine the true shear rate at the wall as

g_ w ¼


 3þb g_ a s1 4

(6)

b is obtained by derivation of the apparent shear rate vs the wall shear stress on a double logarithmic plot, and represents the slope of the curve on such plot.

b¼

dlogg_ a dlogtw

(7)

The melt viscosity (h) is then calculated through the relation between both corrected shear stress and shear rate as

h¼

2.1. Theoretical principles

59

tw ðPa$sÞ g_ w

(8)

Lab Kars software from Alpha Technologies [24] can be used to determine the true shear stress, the true shear rate and the melt viscosity, using capillary rheological data acquired over a 10 to 6000 (s1) apparent shear rate range. The melt viscosity values for each experiment correspond to the average of at least 3 trials. 2.2. Rheological behaviour modelling Several constitutive models have been developed to predict the rheological behaviour of thermoplastic materials for injection moulding purposes, including Newtonian, Ostwald (Power-Law), Cross and Carreau models, all of them relating viscosity to other parameters, such as temperature, shear rate and pressure [25]. The Cross-WLF model is based on the equation:

60

P. Lima et al. / Polymer Testing 45 (2015) 58e67

h¼ 1þ




h0 h0 t*

g

ð1nÞ

(9)

h0 (Pa.s) e reference viscosity under zero-shear-rate conditions; t* (Pa) e model constant which gives the shear stress at the transition between the Newtonian and non-Newtonian behaviour, i.e. the starting point for the pseudo-plastic behaviour of the material, n e power law index. This equation is extensively used for the injection moulding simulation. It describes the Newtonian and shear thinning flow regions, providing a good approximation to the real processing behaviour of the thermoplastic materials during the filling and post filling stage [26]. The viscosity of the material under zero-shear-rate condition can be determined by the WilliamseLandeleFerry expression [27]. A1 ðTT~ Þ ~

h0 ¼ D1 $eA2 þðTT Þ if T  T~

(10)

with:

~ þ D $p A2 ¼ A 2 3 T~ ¼ D2 þ D3 $p where Te(K) is the transition temperature of the material, dependent on the pressure. D2 (K) is the model constant related to the transition temperature of the material at atmospheric pressure. D3 (K/Pa) is the model constant which includes the material temperature variation, as a function of pressure. D1 (Pa s1) is the model constant that gives the material viscosity under zero shear rate conditions, at the transition temperature and atmospheric pressure. A1 is the model constant that includes the temperature dependence of the transition temperature under zero shear rate ~ 2 (K) is the WLF parameter that depends on the type of conditions. A material. The Cross-WLF parameters can be classified as universal con~ 2), related to the intrinsic material behaviour, stants (D2, D3, and A and the other four (n, t*, D1, A1) dependent of each material grade. 3. Experimental 3.1. Materials The rubber materials used in this study are as follows: GTR from mechanical ground scrap tyres, FB 00-08, from Biosafe S.A., Portugal, obtained by an ambient grinding process, sieving class 635 to 20 mesh and density from 0.6 to 0.7 g/cm3; an EDPM rubber, Buna® EP G2470 from Lanxess, with 68 wt% ethylene content, 4.2 wt% content of ethylidiene norbornene (ENB) as diene, 0.86 g/ cm3 density and a melt flow index (MFI) of 0.5 g/10 min (230
C, 2.16 kg); and the EPR, Vistamaxx™ 6202, from Exxon Mobile, with 15 wt% ethylene content, density of 0.86 g/cm3 and a MFI of 26 g/ 10 min (230
C, 2.16 kg). A polypropylene homopolymer, PPH10060 supplied by Total Petrochemical, suited for injection moulding of very thin and/or complex parts, was used as the thermoplastic material, having a MFI of 35 g/10 min (230
C, 2.16 kg) and density of 0.91 g/cm3. 3.2. Blends composition An experimental procedure was established to evaluate the

individual and combined effects of the rubber materials on the flow behaviour of PP based blends (Table 1). A first set of binary blends, with different weight content, was developed to study the individual effect of the rubber components: EPR, EPDM and GTR. A second set of ternary blends was formulated to study the effect of 30%wt content replacement of fresh rubber material (EPDM or EPR) by GTR. 3.3. Melt mixing and samples preparation The blends preparation was made in a Brabender type mixer at 180
C and 60 rpm rotor speed. For the binary blends formulation, the PP was placed in the mixer chamber and, after a 2 min period, the rubber components were added for an additional 8 min mixing time. The TPEGTR formulation was performed in two stages to achieve better encapsulation of the GTR particles by the fresh rubber material, EPDM or EPR [9]. In the first stage, the fresh rubber and GTR were mixed for 8 min and then granulated for the next blending stage. In the second stage, PP was placed in the mixer chamber for 2 min. The rubber mixtures prepared in the first phase were then added and mixed for an additional 8 min period. Similar experimental mixing conditions were also established in other works without reporting any thermal degradation of the materials [9,28]. The blends were then granulated for injection moulding purposes. In order to produce test specimens, a 65 tons Inauton D65 injection moulding machine was used with the following parameters: 220
C injection temperature, 35 bar holding pressure and mould temperature of 40
C. 3.4. Rheological characterisation The rheological properties of the materials must be obtained to predict the processability of TPEGTR blends. Due to the high shear rates involved in the injection moulding process (over 500 s1), the rheological tests were performed using a dual bore capillary rheometer from Dynisco, model LCR 7002. The rheometer was equipped with two capillary dies with the same diameter (1 mm) but with two different L/D ratios (5 and 30). Rheological data were measured at temperatures of 200
C, 210
C and 220
C for all the blends over a 10 to 6000 (s1) shear rate range. The pressure drop for a given volumetric flow rate or shear rate was measured simultaneously for both dies and the capillary rheological data were subjected to Bagley and Weissenberg-Rabinowitsch corrections to determine the true shear rate (g_ w ) and the true shear stress (tw ). Parameters for each blend were determined by fitting the CrossWLF model to the real viscosity curves obtained by the capillary rheological experiments. The Chi-squared method was used to determine these parameters through the Solver addein program in MS Office Excel. It consists of the minimisation of the objective function

O ¼ min

Pre X X hObs i;j  hi;j i

j

hPre i;j

!2 (11)

hObs and hPre are the observed and predicted viscosities at a i;j i;j given temperature (Ti) and shear rate (g_ j ). which represents the sum of the Chi-square errors across temperature (i) and shear rate (j).The fitting was made setting A2 and D3 as 51.6 K and 0 respectively [27], and allowing the variation of n, t*, D1, D2 and A1. The adjustment between the predicted and observed data was made, obtaining a correlation value close to 1.

P. Lima et al. / Polymer Testing 45 (2015) 58e67

61

Table 1 Blends: nomenclature and composition. Blends composition: PP (P); GTR (G) Rubber (R): EPDM (E) or EPR (V); Px1(Ry1Gy2)x2 (1) Reference material: Binary blends Px1(Ry1Gy2)x2 x1 ¼ (70; 50; 30) y1 ¼ (0; 1)

R ¼ EPDM (E) with: y2 ¼ 0

Px1Gx2y2 with: x1 ¼ 70 and y1 ¼ 0 R ¼ EPDM (E)

R ¼ EPR (V)

3.5. TimeeTemperature Superposition Principle (TTSP) For successfully simulating the injection moulding process, it is important to know the viscoelastic behaviour of the materials under different processing conditions, such as temperature, pressure and injection speed. Due to the time-temperature dependence behaviour of the thermoplastic blends, this would imply obtaining the necessary rheological data for each specific application, resulting in an unrealistic number of rheological tests. To overcome this restriction, the TimeeTemperature Superposition Principle (TTSP) was used to define a single behaviour for each blend under different shear rates and temperatures [29]. TTSP allows the construction of a master curve covering a wide temperature range. To obtain the master curve, the viscosity curves obtained at different temperatures (200, 210 and 220
C) were superposed using a temperature dependent shift factor (aT). The reference temperature was selected as 210
C. The superposition temperature dependent shift factor aT is determined as

aT ðTÞ ¼

PP (%wt)

EPDM (%wt)

EPR (%wt)

GTR (%wt)

P100 P70E30 P50E50 P30E70 P70V30 P50V50 P30V70 P70G30 P70(E0.7G0.3)30 P50(E0.7G0.3)50 P30(E0.7G0.3)70 P70(V0.7G0.3)30 P50(V0.7G0.3)50 P30(V0.7G0.3)70

100 70 50 30 70 50 30 70 70 50 30 70 50 30

0 30 50 70 0 0 0 0 21 35 49 0 0 0

0 0 0 0 30 50 70 0 0 0 0 21 35 49

0 0 0 0 0 0 0 30 9 15 21 9 15 21

Constraints: x1 þ x2 ¼ 100 y1 þ y2 ¼ 1

R ¼ EPR (V) with: y2 ¼ 0

Ternary blends Px1(Ry1Gy2)x2 x1 ¼ (70; 50; 30) y1 ¼ 0.7

Designation

h0 ðTÞ h0 ðT0 Þ

(12)

from which the reduced viscosity (h=aT ) and reduced shear rate _ T ) were obtained: (g$a The master curve for each blend was then fitted through the Cross-WLF model and the master curve Cross-WLF coefficients determined. 4. Results and discussion 4.1. Rheological behaviour PP, EPR and EPDM The rheological behaviour of PP, EPR and EPDM at 220
C is shown in Fig. 1. This temperature was selected in order for it to be possible to obtain all the rheological data for EPDM. Viscosity of EPDM decreases almost linearly with increasing shear rate, which means non-Newtonian flow behaviour is observed across nearly the entire range. A higher entanglement of the EPDM molecular chains can lead to more pronounced elastic behaviour, resulting in higher orientation of the EPDM chains under flow. On the other hand, PP and EPR show a viscosity plateau at low shear rates, characteristic of Newtonian behaviour, and non-Newtonian behaviour at higher shear rates. At low shear rates, the entanglements between the chains are able to open by Brownian motions,

sliding past each other, which results in Newtonian flow behaviour [30,31]. At high shear rates, the entangled chains do not have the necessary time to disentangle and are forced to orientate along the flow, resulting in decrease of flow resistance and, thereby, shear thinning behaviour. The rheological parameters determined by the Cross-WLF model (Table 2) allow us to support the observations made from the experimental results. The higher values of t* reveal that EPR has a significantly greater Newtonian plateau than EPDM and PP. A smaller n value for EPR also shows its higher shear thinning behaviour. EPR has a considerably lower zero shear rate viscosity (h0) than EPDM. This rheological behaviour of EPR means that, besides its proven rubber toughening effect on PP based blends [19], it also has the appropriate characteristics to be used on the development of blends for injection moulding of thin and/or complex parts. 4.2. Rheological behaviour of binary and ternary blends All the analysed blends showed a decrease of viscosity with increasing shear rate (Tables 3e6; Figs. 2 and 3). This reveals a shear thinning behaviour that arises from the flow induced shear of entangled and randomly orientated polymer chains. With increasing shear rate, the chain alignment in the flow direction and a higher disentanglement of the macromolecules leads to decrease of the flow resistance and, consequently, of viscosity [32]. The blends exhibit a low power law index (n < 0.3), indicating highly shear thinning behaviour. This characteristic flow behaviour reveals that TPEGTR blends are potentially viable materials for the injection moulding process. As expected, temperature has a significant effect on the rheological behaviour. All the blends tested at the highest temperature (220
C) showed the lowest viscosity. This is a consequence of a higher free volume between molecules with increasing temperature, which leads to reduction of friction between the molecules, and hence of viscosity. Additionally, it is possible to see that the temperature effect on viscosity is stronger at lower shear rates, gradually disappearing with increasing shear rate (see Fig. 2). This indicates that at higher shear rates the effect of the entanglements tends to override the temperature effect. The transition from Newtonian to non-Newtonian behaviour is shown to be temperature dependent. At low shear rates, higher temperatures favour the disentanglement and slippage of the molecular chains and, thereby, enhancing the Newtonian regime [33].

62

P. Lima et al. / Polymer Testing 45 (2015) 58e67

Fig. 1. Viscosity of PP, EPR and EPDM as a function of shear rate.

Table 2 Cross-WLF parameters for PP, EPR and EPDM materials. Cross-WLF parameters

PP (P)

EPR (V)

EPDM (E)

h0(Pa∙s) t* (Pa)

495 26166 0.32

632 96244 0.21

14897 21026 0.40

n

4.2.1. Influence of GTR, EPDM and EPR on viscosity The increase of the blend viscosity with GTR content shows its semi-rigid vulcanized nature, with limited deformation under shear stress. Under these conditions, GTR behaves like a filler

constraining the orientation and disentanglement of the PP molecular chains, thereby increasing the flow resistance, especially at low shear rates [34]. With increasing shear rates, higher deformation of the GTR particles under shear flow leads to lower resistance and, thereby, to similar viscosity values of P70G30 blend and PP. The EPDM presence in the binary blends with 30% weight induces higher viscosity and elasticity than the EPR presence. The shape and size of the elastomer dispersed particles, their molecular weight and degree of entanglements can explain this rheological behaviour. Unlike GTR particles, the dispersed domains of noncrosslinked EPR and EPDM can be extensively changed during flow, deforming under stress field and recoiling afterwards [35]. In

Table 3 Cross-WLF parameters for GTR and EPDM based binary blends. Cross-WLF parameters

h0(Pa∙s) t (Pa) n

P70G30

P70E30

P50E50

P30E70

200
C

210
C

220
C

200
C

210
C

220
C

200
C

210
C

220
C

200
C

210
C

220
C

1186 33344 0.25

970 30689 0.25

723 34963 0.22

1571 27378 0.28

981 28232 0.29

908 25145 0.30

1985 32397 0.29

1523 38046 0.27

1424 32395 0.29

3831 39136 0.27

3443 39364 0.27

2804 36741 0.30

Table 4 Cross-WLF parameters for EPR based binary blends. Cross-WLF parameters

h0(Pa∙s) t (Pa) n

P70V30

P50V50

P30V70

200
C

210
C

220
C

200
C

210
C

220
C

200
C

210
C

220
C

846 33182 0.30

598 33953 0.30

425 39942 0.31

720 49210 0.26

518 47859 0.26

447 51255 0.27

578 78208 0.21

463 74613 0.22

359 65000 0.17

Table 5 Cross-WLF parameters for EPDM based ternary blends. Cross-WLF parameters

h0(Pa∙s) t (Pa) n

P70(E0.7G0.3)30

P50(E0.7G0.3)50

P30(E0.7G0.3)70

200
C

210
C

220
C

200
C

210
C

220
C

200
C

210
C

220
C

1900 26428 0.28

1166 31908 0.26

851 41034 0.22

2478 32841 0.27

2186 30928 0.27

1579 34403 0.25

4785 39446 0.25

3726 37788 0.26

2741 39601 0.26

P. Lima et al. / Polymer Testing 45 (2015) 58e67

63

Table 6 Cross-WLF parameters for ternary EPR based blends. Cross-WLF parameters

h0(Pa∙s) t (Pa) n

P70(V0.7G0.3)30

P50(V0.7G0.3)50

P30(V0.7G0.3)70

200
C

210
C

220
C

200
C

210
C

220
C

200
C

210
C

220
C

891 39377 0.27

720 42429 0.24

478 45457 0.24

763 63259 0.22

648 55644 0.24

561 57709 0.22

1088 68142 0.21

784 72065 0.21

593 78256 0.18

Fig. 2. Viscosity of PP and 70/30 (wt%) blends as function of shear rate, at 200, 210
C and 220
C.

64

P. Lima et al. / Polymer Testing 45 (2015) 58e67

Fig. 3. Viscosity of binary blends with different rubber content (%wt) as function of shear rate, at 220
C.

previous studies [19,20], a continuous-disperse morphology was observed for the 70/30 blends composition, with the rubber particles dispersed in a continuous PP matrix. With this type of morphology, the rheological properties are strongly dependent of PP matrix [31]. The higher viscosity of these blends is mainly attributed to the elastic nature of the rubber dispersed phase, which has a slower response to the induced shear. The P70V30 blend morphology revealed smallest rubber domains (below 1 mm) dispersed in a continuous PP matrix, which could lead to a smaller EPR effect on the blends rheology. Higher EPDM molecular weight and more entangled molecular chains can also lead to this outcome for PP based blends. 4.2.2. Effect of the rubber content on the TPE blends Results also showed an increase of viscosity with increasing EPDM and EPR weight content (Fig. 3). At low rubber content, the rheological behaviour is close to neat PP. With increasing rubber content, EPDM shows a higher effect than EPR that can be explained by the nature of the materials and the blends morphology. In previous works, the morphology of the studied blends was analysed [19,20]. PP/EPDM blends revealed a transition to a co-continuous phase above 50% rubber weight content. This morphology seems to be responsible for a more preponderant effect of EPDM on the rheology of the blend. On the other hand, the EPR blends morphology revealed compatibility between the materials, which can explain the predominance of the PP rheological behaviour at higher EPR contents. The rheological curves generated using the Cross-WLF model (Fig. 3) revealed a particular evolution of viscosity versus shear rate for PP/EPR blends. The blends with higher EPR content have higher viscosity at shear rates above 10 s1, a trend observed from the experimental data obtained by capillary rheometry and also from the data provided by the Cross-WLF model. However, for shear rates below 10 (s1), the Cross-WLF model predicts that blends with higher rubber content exhibit lowest zero-shear rate viscosities (h0). The increase of the Newtonian plateau for higher EPR contents, seen by the increment of t*, combined with a higher shear thinning behaviour, seen by the slope of the curve (1-n), can explain this rheological behaviour [36]. The fact that rheological data were experimentally obtained from the highest to the lowest shear rates can help to understand these results. Transition from shear thinning to Newtonian regime, occurring at a higher shear rate, means faster stabilization of viscosity towards a constant value, resulting

in a lower increment. A transition occurring at lower shear rates means that viscosity increases for a longer period until reaching a constant value. This different evolution of the viscosity with decreasing shear rate induces, beyond a certain value, a shift in the rheological curves position. This behaviour can also be seen in Fig. 1 for the PP and EPR materials. The fitting of the Cross-WLF model to the experimental data may also explain the observed behaviour. As fitting does not include any rheological data for shear rates below 10 (s1), the results predicted for that region may be less accurate. It is noteworthy that values given by the Cross-WLF model for shear rates in the same range as those obtained using the capillary rheometer show good agreement with the experimental data, as can be seen from the curves in Figs. 2 and 3, the correlation factor being very close to 1. 4.3. Master curves Master curves of each blend (Fig. 4) were determined at the intermediate temperature of 210
C, and allowed the determination of the Cross-WLF parameters for each blend (Tables 7 and 8). 4.3.1. Binary blends master curves For the 70/30 weight content binary blends (P70R30), the GTR presence leads to the highest zero shear rate viscosity (h0) and EPDM induces a higher h0 than EPR (Table 7). The h0 values obtained from the master curves fitting showed the same trend as regards the rubber content effects (already observed in Fig. 3): an increase of h0 with increasing EPDM content and a decrease of h0 with the increase of EPR content. 4.3.1.1. Pseudoplastic behaviour. Rubbers induce a larger Newtonian regime in the binary blends, seen by the transition to shear thinning behaviour at higher shear rates (Fig. 4). This is also observed through an increase of t* predicted by the Cross-WLF model (Fig. 5). EPR induces the highest Newtonian regime in the P70R30 blends and EPDM the lowest. This agrees with the individual behaviour of EPDM and EPR observed from the viscosity curves (Fig. 1), where EPDM shows highly shear thinning behaviour. This is a desirable characteristic for injection moulding purposes as viscosity starts to decrease at lower shear rates, implying easier flow and lower energy costs. On the other hand, the higher viscosity of EPDM can be a

P. Lima et al. / Polymer Testing 45 (2015) 58e67

65

Fig. 4. Master curves for TPEGTR blends. Table 7 Cross-WLF parameters for PP and TPE blends. Master curve (210
C) Cross-WLF parameters

P100

P70V30

P50V50

P30V70

P70E30

P50E50

P30E70

P70G30

h0(Pa∙s) t * (Pa)

645 24208 0.32

576 37247 0.30

538 49955 0.25

519 62013 0.24

961 27726 0.29

1533 33962 0.28

3346 39152 0.28

977 32148 0.25

n

Table 8 Cross-WLF parameters for TPEGTR blends. Master curve (210
C) Cross-WLF parameters

P70(V0.7G0.3)30

P50(V0.7G0.3)50

P30(V0.7G0.3)70

P70(E0.7G0.3)30

P50(E0.7G0.3)50

P30(E0.7G0.3)70

h0(Pa∙s) t* (Pa)

663 41146 0.27

658 58140 0.22

787 72321 0.20

1202 31370 0.26

2219 32129 0.26

3642 39790 0.25

n

disadvantage for the injection process of thin or/and complex parts. With increasing rubber content, a wider Newtonian behaviour is observed, which is indicated by an increase of t*, slightly higher shear thinning and general reduction of the power law index, n. 4.3.2. TPEGTR blends master curves The incorporation of GTR in the TPE binary blends introduces new interfacial interactions within the composite, namely with PP

and fresh rubber materials. The strength of these interactions and the rheological behaviour of each of the materials will affect the morphology and rheology of the TPEGTR blends. As reported in previous studies [19,20], these ternary blends present a complete or partial coreeshell morphology of GTR/EPR or GTR/EPDM within a PP/rubber continuous-disperse or co-continuous morphology. The partial substitution of the EPR and EPDM rubber by GTR leads to a general increment of t* and h0 and to a reduction of n

66

P. Lima et al. / Polymer Testing 45 (2015) 58e67

Fig. 5. Cross-WLF parameters as function of rubber weight content (wt%).

(Fig. 5 and Table 8). These results agree with the observed effect of GTR on the P70G30 blend. Thus, higher shear thinning behaviour and wider Newtonian plateau of the TPEGTR blends are observed. Pseudoplastic behaviour is observed for the TPEGTR blends, indicating that GTR can be used in TPE blends without compromising its processability. The interaction of GTR with EPDM or EPR on TPEGTR blends can also be analysed. The increase in viscosity is smaller in the ternary blends with EPR, especially at lower shear rates. Higher compatibility of EPR with PP, due to its high propylene content (above 80%) and the encapsulation of GTR by EPR, can induce higher deformation of GTR under shear flow and consequently lower viscosity. TPEGTR blends based on EPR show a smaller viscosity, close to the pure highly flowable PP, even for the P30(V0.7G0.3)70 blend with 21 wt % GTR. This rheological behaviour shows that formulation of TPEGTR blends for the injection moulding industry is a feasible strategy for GTR recycling, which can broaden the application for this type of potentially wasted material in new and interesting areas.

strategy for the development of TPEGTR blends with adequate flowability for the injection moulding process, counterbalancing the viscosity increase induced by GTR. The incorporation of a rubber component in the ternary blends, such as EPR, with low viscosity, pseudoplastic behaviour and compatible with PP, enables the production of new TPEGTR blends with good processability characteristics for injection moulding applications. The numerical method used for the determination of the CrossWLF rheological parameters provided a very close estimation to the experimental data obtained by capillary rheometry. The Cross-WLF model can be used in a new stage of the TPEGTR industrialization process, supported by numerical simulations and experimental trials of the injection moulding process. This study shows that injection moulding has the potential to be considered for large scale recycling of GTR. The development of TPEGTR blends can be an adequate strategy to upcycle this type of potentially wasted material. References

5. Conclusions The rheology of binary and ternary PP based blends, with EPDM, EPR and GTR as rubber components, was studied, revealing characteristic pseudoplastic behaviour of all the analysed blends. The use of a highly flowable PP material proved to be a suitable

[1] C. Clauzade, Analyse du cycle de vie pour 9 voies de valorisation des pneus ference, R&D Aliapur, juin, 2010. usages non reutilisables - document de re drine, End of life vehicles, in: E.P. Council (Ed.), Directive [2] N. Fontaine, H. Ve 2000/53/CEOfficial Journal of the European Union, 2000, p. 18. szaros, T. B ny, Ground tyre rubber (GTR) in ther[3] J. Karger-Kocsis, L. Me ara moplastics, thermosets, and rubbers, J. Mater. Sci. 48 (2013) 1e38. [4] M. Awang, H. Ismail, Preparation and characterization of polypropylene/waste

P. Lima et al. / Polymer Testing 45 (2015) 58e67

[5]

[6]

[7]

[8]

[9]

[10] [11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

tyre dust blends with addition of DCP and HVA-2 (PP/WTDP-HVA2), Polym. Test. 27 (2008) 321e329. ^meros terM.M. Cossa, A.S. Sirqueira, B.G. Soares, Desenvolvimento de elasto sticos vulcanizados (TPV) a base de polipropileno com resíduo de pneu: mopla I - Planejamento fatorial de experimentos, Polimeros 19 (2009) 190e195. O. Grigoryeva, A. Fainleb, V. Shumskii, V. Vilenskii, N. Kozak, N. Babkina, The effect of multi-reprocessing on the structure and characteristics of thermoplastic elastomers based on recycled polymers, Polym. Sci. Ser. A 51 (2009) 216e225. H. Ismail, M. Awang, M.A. Hazizan, Effect of waste tire dust (WTD) size on the mechanical and morphological properties of polypropylene/waste tire dust (PP/WTD) blends, Polym. Plast. Technol. Eng. 45 (2006) 463e468. C.R. Kumar, I. Fuhrmann, J. Karger-Kocsis, LDPE-based thermoplastic elastomers containing ground tire rubber with and without dynamic curing, Polym. Degrad. Stab. 76 (2002) 137e144. E.J. Lievana, Recycling of Ground Tyre Rubber and Polyolefin Wastes by Producing Thermoplastic Elastomers, Fachbereich für Maschinenbau und Verfahrenstechnik, Technical University of Kaiserslautern, 2005. J. Qin, H. Ding, X. Wang, M. Xie, Z. Yu, Blending LLDPE and ground rubber tires, Polym. Plast. Technol. Eng. 47 (2008) 199e202. S. Wiessner, U. Wagenknecht, M. Zichner, H. Michael, G. Heinrich, Effects of interface reactions in compatibilised ground tyre rubber polypropylene elastomic alloys, Plast. Rubber Compos. 35 (2006) 393e400. S.L. Zhang, Z.X. Zhang, Z.X. Xin, K. Pal, J.K. Kim, Prediction of mechanical properties of polypropylene/waste ground rubber tire powder treated by bitumen composites via uniform design and artificial neural networks, Mater. Des. 31 (2010) 1900e1905. X. Zhang, C. Lu, M. Liang, Preparation of thermoplastic vulcanizates based on waste crosslinked polyethylene and ground tire rubber through dynamic vulcanization, J. Appl. Polym. Sci. 122 (2011) 2110e2120. E. Prut, O. Kuznetsova, J. Karger-Kocsis, D. Solomatin, Rheological properties of ground rubber tire filled isotactic polypropylenes of different molecular weight characteristics, J. Reinf. Plast. Compos. (2012). H.M. Costa, V.D. Ramos, Analysis of thermal properties and rheological behavior of LLDPE/EPDM and LLDPE/EPDM/SRT mixtures, Polym. Test. 27 (2008) 27e34. J.K. Kim, S.H. Lee, M. Balasubramanian, A comparative study of effect of compatibilization agent on untreated and ultrasonically treated waste ground rubber tire and polyolefin blends, Polim. 16 (2006) 263e268. S.L. Zhang, Z.X. Xin, Z.X. Zhang, J.K. Kim, Characterization of the properties of thermoplastic elastomers containing waste rubber tire powder, Waste Manage 29 (2009) 1480e1485. R. Scaffaro, N.T. Dintcheva, M.A. Nocilla, F.P. La Mantia, Formulation, characterization and optimization of the processing condition of blends of recycled polyethylene and ground tyre rubber: mechanical and rheological analysis, Polym. Degrad. Stab. 90 (2005) 281e287. P.S. Lima, J.M. Oliveira, V.A.F. Costa, Partial Replacement of EPR by GTR in

[20]

[21] [22]

[23] [24] [25] [26] [27]

[28]

[29] [30] [31]

[32]

[33] [34]

[35] [36]

67

highly flowable PP/EPR blends: effects on morphology and mechanical properties, J. Appl. Polym. Sci. (2015) 42011. P.S. Lima, J.M. Oliveira, V.A.F. Costa, Partial replacement of EPDM by GTR in thermoplastic elastomers based on PP/EPDM: effects on morphology and mechanical properties, J. Appl. Polym. Sci. 131 (2014) 40160. E.B. Bagley, End corrections in the capillary flow of polyethylene, J. Appl. Phys. 28 (1957) 624e627. C. Nakason, S. Jamjinno, A. Kaesaman, S. Kiatkamjornwong, Thermoplastic elastomer based on high-density polyethylene/natural rubber blends: rheological, thermal, and morphological properties, Polym. Adv. Technol. 19 (2008) 85e98. R. Eisenschitz, B. Rabinowitsch, K. Weissenberg, Mitteil deutsch, Materialspruefungsamt, Sonderheft 9 (1929) 91. Capillary Rheometer (LCR 7000 Series), http://www.dynisco.com/capillaryrheometerelcr-7000-series-. P. Kennedy, R. Zheng, Flow Analysis of Injection Molds, Carl Hanser Verlag, 2013. M.M. Cross, Rheology of non-newtonian fluids: a new flow equation for pseudo-plastic systems, J. Colloid Sci. 20 (1965) 417e437. M.L. Williams, R.F. Landel, J.D. Ferry, The temperature dependence of relaxation mechanisms in amorphous polymers and other glass-forming liquids, J. Am. Chem. Soc. 77 (1955) 3701e3707. O.P. Grigoryeva, A.M. Fainleib, A.L. Tolstov, O.M. Starostenko, E. Lievana, J. Karger-Kocsis, Thermoplastic elastomers based on recycled high-density polyethylene, ethylene-propylene-diene monomer rubber, and ground tire rubber, J. Appl. Polym. Sci. 95 (2005) 659e671. J. Dealy, D. Plazek, Time-temperature superpositionda users guide, Rheol. Bull. 78 (2009) 16e31. J.D. Ferry, Viscoelastic Properties of Polymers, John Wiley & Sons, 1980. O. Balkan, A. Ezdes¸ir, Rheological behaviors of glass bead- and wollastonitefilled polypropylene composites modified with thermoplastic elastomers, Polym. Compos 33 (2012) 1162e1187. L. D'Orazio, C. Mancarella, E. Martuscelli, G. Sticotti, G. Cecchin, Isotactic polypropylene/ethylene-co-propylene blends: influence of the copolymer microstructure on rheology, morphology, and properties of injection-molded samples, J. Appl. Polym. Sci. 72 (1999) 701e719. R.K. Gupta, Polymer and Composite Rheology, CRC Press, 2000. Z. Li, S. Guo, W. Song, Y. Yan, Effect of the interfacial interaction on the phase structure and rheological behavior of polypropylene/ethylenee octene copolymer/BaSO4 ternary composites, J. Polym. Sci. Part B Polym. Phys. 40 (2002) 1804e1812. M. Gahleitner, Melt rheology of polyolefins, Prog. Polym. Sci. 26 (2001) 895e944. , R. Navarro, Study of rheological, thermal, J.E. Crespo, F. Parres, M.A. Peydro and mechanical behavior of reprocessed polyamide 6, Polym. Eng. Sci. 53 (2013) 679e688.