1. utilization of tire rubber and recycled polyolefins into

T Transworld Research Network 37/661 (2), Fort P.O. Trivandrum-695 023 Kerala, India

Recent Developments in Polymer Recycling, 2011: 1-46 ISBN: 978-81-7895-524-7 Editors: A. Fainleib and O.Grigoryeva

1. Utilization of tire rubber and recycled polyolefins into thermoplastic elastomers

Alexander Fainleib1, Olga Grigoryeva1, Boulos Youssef2,3

and Jean-Marc Saiter2 1Institute of Macromolecular Chemistry of the National Academy of Sciences of Ukraine

Kharkivske shose, 48, 02160 Kyiv, Ukraine; 2AMME-LECAP International Laboratory, EA4528 Institut des Matériaux, Facultés des Sciences, Université de Rouen, BP 12, 76801

Saint Etienne du Rouvray, Cedex, France; 3Institut National des Sciences Appliquées Avenue 76801 Saint Etienne du Rouvray, Cedex, France

Abstract. In the review the methods of producing thermoplastic elastomers with benefit properties from post-consumer polyethylenes (HDPE, LDPE) and recycled rubbers (ground tire rubber, GTR) using several approaches of their pre-treatment and compatibilization procedures are discussed. Some additions of reactive polyethylenes and rubbers were used for improvement of interface adhesion in the blends studied. The TPE produced were characterized by TGA, DSC, DMTA, rheology measurements, X-ray diffraction and mechanical testing. For all of TPE studied the increasing components compatibility due to the formation of the essential interface layer has been observed. The method of renewal of GTR and its reactivation using its high temperature treatment with bitumen has been developed and successfully applied. High performance thermoplastic elastomers (TPEs) based on post-consumer high-density polyethylene (HDPE-pc), low density polyethylene (LDPE-pc), butadiene rubber (BR), olefinic type

Correspondence/Reprint request: Dr. Alexander Fainleib, Institute of Macromolecular Chemistry of the National Academy of Sciences of Ukraine, Kharkivske shose, 48, 02160 Kyiv, Ukraine. E-mail: [email protected]

Alexander Fainleib et al. 2

ethylene/propylene/diene monomers (EPDM) containing rubber and ground tire rubber (GTR) treated with bitumen have been prepared by using the dynamic vulcanization technology, and their structure-property relationships were investigated. It was established that at special pre-treatment of GTR by bitumen bestows on the resulting TPEs outstanding mechanical properties. SEM, DSC and DMTA results revealed an improved adhesion between the particles of GTR treated by bitumen and a surrounding thermoplastic matrix compared to the untreated GTR particles. GTR containing TPEs prepared by extrusion technology were reprocessed (by passing through the extruder 6 times) without noting any significant change in their tensile properties, thermal stability and melt viscosity. 1. Introduction Thermoplastic elastomers (TPEs), especially blends of elastomer and thermoplastic obtained by dynamic vulcanization of rubber in thermoplastic and having characteristics of elastomers while maintaining the thermoplasticity are under serious interest of scientists and producers last decade [1-5]. In a point of view of both economical and ecological reason the replacement of virgin components of TPEs (fully or partly) by recycled polymers is very important. The problem is to obtain materials of beneficial properties (preferably, not inferior to traditional TPEs in main properties). Waste plastics, especially polyolefins, and rubbers, including tire rubber, have caused a series of environmental problems. Many approaches have been proposed to use the large amount of waste polymers. The standard use is a replacement of a part of virgin polyolefin, for example polyethylene, by some recycled grades. Similarly, part of virgin rubber is replaced by ground tire rubber (GTR) in less demanding and even in tire formulations [6, 7]. In recent years, a potential way to use GTR in thermoplastic elastomers has been developed. Numerous investigations [8-17] including our own ones have shown that introducing GTR directly into recipes of different polyolefin/rubber TPEs results in drastic decreasing their tensile strength and especially ultimate elongation. This is the result of poor interphase adhesion between the blend components [18, 19]. Various modifiers were used to compatibilize rubber/polyolefin blends with and without reclaimed GTR. A short review about the advantages of functionalization and compatibilization of TPEs including GTR has been recently done by Li et al. [20, 21]. For reactive compatibilization (it seems to be the most effective method), the components of TPEs including GTR should be functionalized or at least their surface has to be activated. It can be done by chemical grafting of reactive monomers onto the polymer surface or, in the case of GTR, by thermal, thermo-mechanical, thermo-chemical, ultrasonic devulcanization techniques, etc. [6, 7, 11, 20].

Utilization of tire rubber and recycled polyolefins into thermoplastic elastomers 3

2. Comparative characterization of virgin and post-consumer polyolefins In order to predict properties of thermoplastic elastomers prepared from recycled polyolefins and rubbers the following virgin and post-consumer polymers were analyzed and their main characteristics were compared [22, 23]: √ Low density polyethylene (LDPE), trademark "Riblene" FC30 (Polimeri

Europe, Rome, Italy), with the following characteristics: Mn=31100, Mw=179200, Mz=487200, Mw/Mn=5.76, Mz/Mw=2.72; Tm=112 oC, Young's modulus, E=220 MPa, tensile strength, TS=18 MPa and elongation at break, EB=600%, MFI190/2.16=0.28 g/10 min and MFI230/2.16=0.8 g/10 min.

√ Linear low density polyethylene (LDPEL), trademark “Lupolen 1840 H” (BASF AG, Ludwigshafen, Germany), MFI190/2.16=1.5 dg/min.

√ Post-consumer low density polyethylene (LDPE-pc) made from greenhouse films of composition LDPE 65-70%, LDPEL 12-17%, EVA copolymer 12-15%, Tm=109 oC, E=180 MPa, TS=16 MPa, EB=500%, MFI190/2.16=0,29 g/10 min and MFI230/2.16=0,95 g/10 min. Post-consumer greenhouse films have been collected in the province of Ragusa (Sicily, Italy) after near one year of exploitation. Post-consumer films have been washed, dried and cut to pieces by an industrial scale.

√ Isotactic polypropylene (PP), trademark “PP-169” (LisichanskNefteOrgsintez, Lisichansk, Ukraine), isotactic index 96%, Mn=150,000, Mw=1,000,000, Mw/Mn=6,7, Young's modulus E=130 MPa, TS=3.7 MPa and EB=530%, MFI190/2.16=3.4 g/10 min.

√ Post-consumer polypropylene (PP-pc), (Roksana Ltd., Kyiv, Ukraine) made from post-consumer packages collected in Kiev (Ukraine), MFI230/2.16=1.9 g/10 min.

√ High-density polyethylene (HDPE), trademark “HDPE 277-73” (KazanOrgSintez, Kazan, Russia), MFI190/5.0=20.6 g/10 min, density 0.961 g/cm3 (at 20 oC).

√ Post-consumer high-density polyethylene (HDPE-pc), MFI190/2.16=2.13 g/10 min, TS= 17.7 MPa and EB=10 %. HDPE-pc made of post-consumer bottle transportation crates collected in Kyiv (Roksana Ltd., Kyiv, Ukraine). Waste of bottle transportation crates were washed, dried and cut to pieces by an industrial apparatus.

The comparison of main characteristics of virgin and post-consumer polyolefins, namely HDPE/HDPE-pc, LDPE/LDPE-pc and PP/PP-pc, was


fulfilled [22, 23] using Thermogravimetric Analysis (DTA, DTG, TG), Wide-angle X-Ray Scattering (WAXS), Dynamic Mechanical Thermal Analysis (DMTA), Differential Scanning Calorimetry (DSC) measurements and mechanical testing. Melt flow index for above mentioned polymers was also determined. The data obtained are summarized in Table 1. The corresponding DTA, DTG and TG curves LDPE and LDPE-pc are given in Figure 1. One can see that the curves (Figure 1a-c) of LDPE and LDPE-pc studied are very similar. TGA curves (Figure 1a) are characterized by presence of one endothermic peak as a result of melting LDPE and LDPE-pc (at 393 K and 388 K, respectively), and a few low-resolved high temperature exothermic peaks due to oxidative destruction of polyethylenes studied. Both LDPE and LDPE-pc have very close temperature of the beginning intensive decomposition near 600 K and char residue values of 3.5 and 5 %, respectively (see Figure 1c). Appearance of additional degradation stage (at 548-693 K) (Figure 1b) and high temperature shift of all TGA, DTG and TG curves, as well as increasing the melting temperature (see Figure 1a) and value of char residue (see Table 2) reflect existence of thermally more stable structures in LDPE-pc in comparison with LDPE. Obviously it is a result of partial degradation of LDPE-pc chains and formation of branched and/or cross-linked polymer chains. WAXS curves of LDPE and LDPE-pc are given in Figure 2a. Both diffractograms show two sharp peaks located at scattering angles of 21.1о and 23.4о (characteristic for orthorhombic crystal cell of polyethylene) identified as the (110) and (200) polyethylene strongest reflections, correspondingly [24]. The diffuse maximum located at 19.5о which corresponds to LDPE-pc amorphous phase scattering maximum [25]. Thus, one can see that there are

Table 1. Properties of virgin and post-consumer polyolefins used.

Melting temperature,

Tm, oC

Degree of crystallinity,

Xc, %

Material

DSC DTA DSC WAXS

Tensile strength,

TS, MPa

Elongation at break, EB, %

Flow activation energy, Ea,

k/mol

Melt flow index,

MFI190/2.16, g/10 min

Gel contenta),

wt.%

HDPE HDPE-pc

135 135

126 115

62 70

58.6 55.6

19.0 17.7

11 8

29.3 33.0

2.02 2.13

0 0

LDPE LDPE-pc

118 116

120 115

47 43

29.2 27.5

11.0 7.1

668 440

47.4 47.4

0.27 0.28

0 6

PP PP-pc

174 174

165 164

52 48

51.5 40.2

15.3 10.5

484 405

47.9 38.3

3.4 1.9 b)

0 0

a) determined as o-xylene insoluble fraction; b)measured at 230 oC.


Table 2. Thermal behavior of LDPE and LDPE-pc.

Stages interval of weight loss, Tonset/Tend, K

Tmax rate, K Weight loss, %

LDPE 463 / 498 478 -2 503 / 573 553 2 673 / 773 743 70 753 / 833 803 91

Char residue, % : 3.5 LDPE-pc

463 / 498 483 -2 498 / 568 533 2 548 / 693 633 10 683 / 773 748 75 773 / 833 813 95

Char residue, % : 5

300 450 600 750 900

Exo

393

388

a)

300 450 600 750 900-1,0

-0,8

-0,6

-0,4

-0,2

0,0

T em perature, K T em perature, K

b )

T em perature, K

Δ m , %Δ m τ -1, % m in -1

300 450 600 750 900

-100

-80

-60

-40

-20

0

c)

Figure 1. Thermogravimetric analysis curves for LDPE (open circle) and LDPE-pc (solid circle): (a) Differential Thermal Analysis (DTA); (b) Differential Thermal Gravimetry (DTG); (c) Thermogravimetry (TG). no appreciable differences in a crystal cell or amorphous phase periodicities because both the LDPE have the same mean size of microcrystals (D)=10.7 and 11.1 nm, respectively, and identical crystal lattice spacing (d)=0.421 nm. The data are summarized in Table 3.


5 10 15 20 25 30 35 40

0

30

60

90

19.5o

(a)

I /10

00 p

ulse

s

2Θ , degree

23.4o

21.1o

LDPELDPE-pc

I /10

00 p

ulse

s

2Θ , degree

5 10 15 20 25 30 35 40

3

6

9 24.320.3

(b)

36.4o

34.6o

31.9o

Figure 2. WAXS curves for: (a) LDPE (open circle) and LDPE-pc (solid circle); (b) BR cured.

Table 3. Properties of LDPE and LDPE.

Tm, K Degree of crystallinity,

X, %

Material

DSC DTA DSC WAXS

Size of crystallites,

D, Å

Crystal lattice

spacing, d, Å

TS, MPa

EB, %

Gel content a),

wt.%

LDPE LDPE-pc

391 389

393 388

47 43

29.2 27.5

10.7

11.1

0.421

0.421

11.0 7.1

668 440

0 6

a)Determined as o-xylene insoluble fraction

However, it can be seen that LDPE has a higher value of degree of crystallinity <X> than LDPE-pc. It can be explained by the reduction of molecular weight of LDPE-pc due to additional thermooxidative destruction as well as crosslinking occurred at their outdoor exploiting and reprocessing. It is clear, that LDPE-pc has higher content of amorphous phase than LDPE. WAXS curves of LDPE and LDPE-pc are given in Figure 2a. Both diffractograms show two sharp peaks located at scattering angles of 21.1о and 23.4о (characteristic for orthorhombic crystal cell of polyethylene) identified as the (110) and (200) polyethylene strongest reflections, correspondingly [24]. The diffuse maximum located at 19.5о which corresponds to LDPE-pc amorphous phase scattering maximum [25]. Thus, one can see that there are


no appreciable differences in a crystal cell or amorphous phase periodicities because both the LDPE have the same mean size of microcrystals (D)=10.7 and 11.1 nm, respectively, and identical crystal lattice spacing (d)=0.421 nm. The data are summarized in Table 3. However, it can be seen that LDPE has a higher value of degree of crystallinity <X> than LDPE-pc. It can be explained by the reduction of molecular weight of LDPE-pc due to additional thermooxidative destruction as well as crosslinking occurred at their outdoor exploiting and reprocessing. It is clear, that LDPE-pc has higher content of amorphous phase than LDPE. The Cp (T) plots of both the LDPEs are shown in Figure 3. One can see that both LDPE and LDPE-pc have typical curves for semicrystalline polyolefins with a phase transition “solid – liquid” in the temperature region 310–391 K and 322–389 K, respectively. Note that the above melting temperature interval and melting peak temperature, Tm=389 K, for LDPE-pc studied are quite similar to the reported values for other post-consumer LDPEs [26]. In contrary to LDPE melting process of crystalline phase of LDPE-pc consists of melting low molecular weight crystallites (probably defected) in the temperature range 322-338 K and melting high molecular weight crystallites at 389 K [27]. We consider that shoulder in the temperature region of 338–350 K without any visible changes in values of Cp relates to a recrystallization of above low molecular weight crystallites of LDPE-pc into the high molecular weight crystallites. The common decrease of values of Cp in temperature region between 322 and 373 K observed for LDPE-pc sample in comparison with LDPE evidences of increasing packing density of polymer due to formation of above mentioned branched or cross-linked polymer chains. It can be seen that LDPE has a higher value of degree of crystallinity <X> than LDPE-pc that is agreed to WAXS data.

200 250 300 350 400 4500,5

1,0

1,5

2,0

2,5

3,0

a)

203

249

319

383

Temperature, K

Cp, J

g-1 K

-1

275 300 325 350 375 400 425

5,0

5,5

6,0

6,5389

b)

Temperature, K

350

310

338

322

391

373

Temperature, K

Cp, J

g-1 K

-1

Figure 3. Temperature dependence of specific heat capacity (Cp) of: (a) BR cured; (b) LDPE (open circle) and LDPE-pc (solid circle).


Tensile properties and residual gel content values for LDPE and LDPE-pc samples are presented in Table 1. Increasing residual gel content value and some reduction in tensile properties observed for LDPE-pc confirm the above conclusions about some branching and crosslinking of LDPE-pc chains. Rheological behavior for LDPE and LDPE-pc is shown in Figure 4. One can see that for both LDPE and LDPE-pc at each temperature studied the flow curves are very similar that evidences of close values of viscosity of the polymers investigated. The shear rate dependence of melt viscosity (in Arrenius coordinates) of both LDPE and LDPE-pc is shown in Figure 5. Rheological behavior of all virgin and post-consumer polyolefins studied is presented graphically in Figure 5. Based on the data presented the flow activation energies, E, were calculated. The data are presented in Table 1.

-2 -1 0 13

4

5

log

τw, P

a

473 K453 K433 K413 K

log γw, s-1

Figure 4. Dependence of shear stress (τw) vs. shear rate (γw) for LDPE (open circle) and LDPE-pc (solid circle) at different temperatures.

-2 -1 0 1

4

5

6log η, Pa slog η, Pa s

473 K

453 K

433 K

log η, Pa s

log γw, s-1log γw, s-1log γw, s-1

LDPE

413 K

-1 0 1 2

3

4

5

473 K453 K

433 K413 K

HDPE

-1 0 13

4

5

493 K

473 K

453 K

PP

Figure 5. Dependence of shear viscosity (η) vs. shear rate (γw) for virgin (open symbols) and post-consumer (solid symbols) LDPE, HDP and PP at different temperatures.


3. Reclamation of ground tire rubber The thermo-mechanical method of reclamation (partial devulcanization) and thus activation of GTR (or at least the surface of GTR particles) has been investigated by using a Brabender Plasticorder (model PLE 330) and different treatment conditions [23]. GTR powder (SCANRUB AS, Viborg, Denmark) of high quality large surface/diameter ratio was produced by grinding in an airgab at supersonic speeds. The main characteristics are given in Table 4. As the powder and granulate is made from a large number of different tire types Genan cannot give any extract values for the elastomeric composition of the powder/Granulate. The following can be used as a guideline: Natural rubber ~30 %, SBR (Styrene-butadiene rubber) ~40 %, BR (butadiene rubber) ~20 %, IIR/XIIR (butyl- and halogenated butyl rubber) ~10 %. CR3 has performed the sieve analysis of 2 fractions of GTR powder used and obtained the following results: fraction 0.4<d<0.7 mm (GTR): d<0.2 mm (0.8 %), 0.2<d<0.4 mm (34.6 %), 0.4<d<0.63 mm (53.4 %), and 0.63<d<1.0 mm (11.2 %). The content of sol fraction was 11.4-12.5% (in acetone according to CR3, CR2 and CR4) and 3.1 % determined in toluene. Based on the analysis of the mixing torque curves (here not reported) it was concluded that the devulcanization of GTR is negligible at relatively low temperatures (180-240 oC) and becomes significant only at high temperatures (300 oC). The sol-gel analysis of processed GTR has shown increasing sol-fraction content from ~12.5% for unprocessed GTR to ~16.1% for GTR processed at 240 oC and up to ~21.4% for GTR processed at 300 oC. The final evidence of the thermal activation of the devulcanization process has

Table 4. Basic characteristics of GTR.

Characteristic Standard Value Specific gravity (g/cm3) ISO 2781 1.1÷1.20

Ash (%) ASTM E1131-86 0÷5 Acetone extract (%) ISO 1407-81 11÷17 Carbon black content* (%) ASTM E 1131-86 32÷36 Rubber hydrocarbon content (%) Calculated ~42 Free textile <1.0 mm (wt. %) Genan 0÷0.8 Free textile >1.0 mm (wt. %) Genan 0÷1.2 Tensile Strength, TS (MPa) ISO R 37 >5.0 Elongation at break, EB (%) ISO R 37 >90 Hardness, Shore A ISO R 868 72 Density (g/cm3) ISO R 868 1.18 Resiliency (%) DIN 53.512 40 Abrasion Loss (mm3) DIN 53.516 220


been confirmed by the measurement of a permanent deformation on materials compressed at a given temperature for 24 hours (ASTM D395/B). The relative results are reported in Table 5. Again, while for the 180 °C processed material no significant differences with the unprocessed GTR can be evidenced, some slight difference is present in the 210 °C and 240 °C processes ones. As for the highest temperature used, in this case no measurement was possible as the material after testing has no consistence and no deformation was measurable. Again, this is a confirmation that relevant changes in the GTR happen only at high temperatures when it is possible to achieve an effective devulcanization of the material. The thermo-chemical method of reclamation (devulcanization) of GTR was also developed. GTR was swollen in a mixture of naphtenic hydrocarbons (industrial processing oil) followed by mixing of swollen GTR with other ingredients (rosin, mineral rubber, indene-coumarone, etc.) by rolls, thermal treatment (T=100-150 oC for 1-10 hours) and second rolling. For the basic GTR and the devulcanized one in regenerate product we have reached an increase of acetone extract value from 12.5 to 18.5 %. These 6 % decrease determines the degree of GTR devulcanization, this fact was confirmed additionally by FTIR data. The molecular mass distribution (MMD) by Size Exclusion Chromatography for acetone soluble fractions of GTR before and after reclaiming has been studied [23]. It was found that the values of molecular masses (Mw, Mn, Mz) of the reclaimed GTR were lower by ~28-43% compared to the initial one. Next considerable efforts were undertaken to find a straightforward method to a better devulcanization of GTR. It was namely suggested that better surface decomposition of GTR allowed a better decomposition of its particles, which can thus participate in the load transfer/dissemination processes during loading and a premature failure of the specimens can be avoided. The concept was to find an ingredient, which may react with sulphur radicals formed due to the break-up of the di-and multi sulphide crosslinks. It was found that bitumen may fulfill this role [28]. Following this presumption a detailed study was devoted to the GTR devulcanization (reclaiming) in bitumen [28-30]. Note that bitumen not only “absorbs” sulphur, it is also a good plasticizer/compatibilizer for various polyolefin blends. The thermo-chemical method of GTR reclaiming using bitumen as a softening and devulcanizing agent was mainly applied at various processing conditions. The acetone soluble fraction due to bituminous reclamation, depending on the processing conditions, was markedly increased by 18 %. It was presumed that the missing ductility (elongation at break) of the thermoplastic compositions can be enhanced and the ductility requirement for the thermoplastic elastomers can be reached when using a bituminous reclamation for GTR


Table 5. Deformation measured by compression set of GTR processed at different conditions.

Residual deformation, % Processing conditions at 23 °C at 70 °C at 100 °C

GTR as it is 6.90 7.41 10.53 160 °C, 5 min, 100 rpm 8.96 10.61 11.76 180 °C, 5 min, 100 rpm 9.49 10.87 11.76 180 °C, 15 min, 100 rpm 10.42 11.27 12.06 210 °C, 5 min, 100 rpm 11.35 12.86 14.69 210 °C, 5 min, 30 rpm 13.99 14.34 15.97 240 °C, 15 min, 100 rpm 15.75 15.07 not measurable 300 °C, 15 min, 100 rpm not measurable not measurable not measurable

Table 6. Characteristics of some bitumens used.

Characteristics Bitumen EN

70/100 Bitumen BN-4 Bitumen BN-5

Softening temperature (oC) Penetration (dmm)

Flash Point (oC) Fire point (oC)

Density at 20oC (g/ml) Asphaltenes content

(wt.%)

50 45 - - -

18÷20

78 30

>240 >360 0.957 31÷32

89 11

>240 >360

1.0091 38÷41

Molar mass distribution a): Mw; Mn;

Mz; Mw/Mn

9,324; 5,516; 17,719; 1,69

12,059; 5,647;

26,491; 2.14

13,923; 6,488; 29,436; 2.15

Element content (wt.%) b): С (carbon) Н (hydrogen)

S (sulfur) N (nitrogen)

Other elements

84.6÷84.82 9.69÷9.56 3.59÷3.39 1.47÷1.42 1.03÷0.43

83.78÷83.90 10.04÷10.27

3.47÷3.57 1.01÷1.20 1.06÷1.7

83.97÷84.07 10.54÷10.72 3.44÷3.62 1.05÷0.92 0.54÷1.13

a) determined by CR4 using Size Exclusion Chromatography method; b) determined by CR4 using elementary analysis method [Mazor L. Methods of organic analysis. Akadémiai Kiadó, Budapest 1983].

without sacrificing the set properties. Some characteristics of the bitumens used are shown in Table 6. The step-by-step (discontinuous) and continuous methods of bituminous treatment of GTR were developed [23, 28] and the influence of processing parameters (temperature, time, screw speed, component content, etc.) on properties of GTR containing final TPEs were studied. The step-by-step


method consisted of such stages: blending of GTR with bitumen, pre-heating the mixture in oven (170 oC for 4-5 hours) followed by rolling of GTR/bitumen blend (for 20-80 min). Finally, different GTR/bitumen containing TPEs have been produced by mastication in the Brabender Plasticorder (model PL 2000) at different processing conditions. The continuous method of bituminous treatment of GTR was performed by using a corotating laboratory twin-screw extruder (Brabender DSE 25). First GTR powder was passed trough the extruder, afterwards, the extruded GTR powder was extruded again together with the bitumen and finally TPE compositions (of different recipe) were produced in the 3rd extrusion run at different processing conditions. The same continuous method of bituminous treatment of GTR was applied by using one-screw extruder (model PLV 150) [28]. First the GRT/bitumen blend was extruded followed by granulation of the extrudate. Afterwards, the granulated GTR/bitumen blend was extruded again together with other components of TPEs. The producing and characterization of TPEs from recycled polyolefins and GTR (or GTR/bitumen blends) will be discussed in Sections 5-9. 4. Reactive compatibilization of components in thermoplastic elastomers based on recycled polyolefins Authors [22, 31] described the method of reactive compatibilization of the recycled LDPE-pc with polybutadiene rubber. The reactive compatibilization method used in this work was realized by introduction of reactive polyethylene copolymer into thermoplastic phase and reactive polybutadiene rubber into the rubber phase to promote the interfacial adhesion by means of chemical interaction between the functional groups of compatibilizing agents in thermoplastic/rubber interface. The chemical reactions occur during melt mixing of components at TPE formation. Schemes of reactions between reactive polyethylenes and reactive polybutadiene rubbers used are shown in Figure 6. Thus, the reactive couples of functional groups can be given as follows: epoxy–carboxyl (1, 2), amine–epoxy (3), amine–anhydride (4), isocyanate–anhydride (5), isocyanate–epoxy (6), isocyanate–carboxyl (7, 8). Tensile properties of TPEs obtained by using different reactive couples are given in Figure 7. One can see that the effect of compatibilization is observed for TPEs obtained by using following reactive couples: PB-NH2/PE-co-GMA, PB-NCO/PE-co-AA and PB-NCO/ PE-co-VA-co-AA. These results indicate that above reactive couples may act as “interfacial agents” promoting adhesion between matrix and dispersed phase in TPEs studied.


Figure 6. Schemes of reactions between reactive polyethylenes and reactive butadiene rubbers used: PB-E/PE-co-AA (1); PB-COOH/PE-co-GMA (2); PB-NH2/PE-co-GMA (3); PB-NH2/PE-g-MAH (4); PB-NCO/PE-g-MAH (5); PB-NCO/PE-co-GMA (6); PB-NCO/PE-co-AA (7); PB-NCO/PE-co-VA-co-AA (8). The highest increase in mechanical characteristics is fixed for LDPE-pc (PE-co-AA)/ BR (PB-NCO) TPE, its values of <TS> and <EB> are higher by 31 % and 63 %, respectively, than for the non-modified LDPE-pc/BR TPE. Undoubtedly, this evidences of effective compatibilization of rubber and polyolefin components in the modified TPE in comparison with non-modified one due to reaction between PE-co-AA and PB-NCO in polyolefin/rubber interface and, as a result, improved interfacial adhesion. This conclusion is confirmed by the data presented in Figures 8-10. One can see that introducing PB-NCO/PE-co-AA (NCO/COOH=1/1) compatibilizer increases the tensile properties of TPE obtained (see Figure 7), which achieves maximal values at ~8¸10 wt.% of PB-NCO (per BR). However, the further increase of PB-NCO content up to 15 wt.% provides a


0

1

2

3

4

5 EB

EB, %

0

87

6543

21

0

100

200

300

400TS, MPa

TS

Figure 7. Tensile properties of unmodified LDPE-pc / BR=60/40 wt.% TPE (0) and the same TPE modified by: PB-E/PE-co-AA (1); PB-COOH/PE-co-GMA (2); PB-NH2/PE-co-GMA (3); PB-NH2/PE-g-MAH (4); PB-NCO/PE-g-MAH (5); PB-NCO/PE-co-GMA (6); PB-NCO/PE-co-AA (7); PB-NCO/PE-co-VA-co-AA (8). All PB-modifiers were used in amount of 7.5 wt.% (per BR) and the ratio of functional groups for PB- and PE-based modifiers was kept 1/1.

Figure 8. Tensile properties of LDPE-pc (PE-co-AA)/BR (PB-NCO) TPEs vs. PB-NCO content in BR phase at NCO/COOH ratio equal 1/1. reduction of EB value. We consider that PB-NCO can also react with unsaturated bonds of the BR inside the rubber phase, which is evident from the gel-content data presented in Figure 9. One can see a non-linear growth of gel content value at increasing PB-NCO content in BR phase. As can be seen from Figure 10 the addition of PE-co-AA to PB-NCO increases the tensile properties of TPEs obtained, which reach a plateau at NCO/COOH=1/1. The excess of PE-co-AA (≥1.5 e.e.w. per PB-NCO) does


Figure 9. Gel content of TPEs studied vs. PB-NCO content in BR phase. LDPE-pc/BR=60/40 wt.% for all TPEs and the ratio of functional groups for PB- and PE-based modifiers was kept 1/1.

0,0 0,5 1,0 1,5 2,01

2

3

4

5

TS

EB

, %

TS,

MPa

PE-co-AA content (per PB-NCO, e.e.w.)in LDPE-pc(PE-co-AA)/BR(PB-NCO)

0

100

200

300

400

EB

Figure 10. Tensile properties of LDPE-pc (PE-co-AA)/BR (PB-NCO) TPEs vs. PE-co-AA e.e.w. per PB-NCO (at 7.5 wt.% PB-NCO content per BR). not influence significantly on tensile properties of the TPEs studied. We explain this by lower reactivity of COOH-groups in comparison with NCO-groups. One can see that minimal values of TS and EB are observed for the TPEs modified by PB-NCO (7.5 wt.% per BR) but without any quantity of PE-co-AA. Clearly, that in this case the interfacial adhesion between the TPE basic components is on the level as for unmodified TPE. In addition, decrease of EB value (by ~15%) for this sample in comparison with unmodified LDPE-pc/BR TPE evidences that PB-NCO indeed participates in reaction with unsaturated bonds of BR inside the rubber phase.


The reason of absence of compatibilizing effect for other reactive couples used is probably connected with kinetic/diffusion peculiarities. One can suppose, that for these couples the reactions of the functional groups in interphase are not so effective and the reagents may be do not have time enough to react completely at the conditions used. LDPE-pc (PE-co-AA)/BR (PB-NCO) TPE has been selected for more detailed investigation of influence of reactive couple content on changing in phase structure, glass transition behavior, degree of crystallinity of polyolefin matrix as well as on thermal and mechanical properties of TPEs studied. 5. Structure-properties relationships for reactively compatibilized thermoplastic elastomers from recycled polyolefins and rubbers Structure-property relationships in LDPE-pc (PE-co-AA) / BR (PB-NCO) TPEs 5.1. Wide-angle X-ray scattering WAXS diffractograms of unmodified and all of modified TPEs studied (Figure 11) show two sharp peaks located at scattering angles of 21.6 and 23.9o (scattering of crystalline phase of polyethylene) as well as the diffuse maximum located at 19.5o (scattering of amorphous phase of polyethylene) [24, 32]. As it was mentioned above three sharp peaks in the region from ~31o to ~36o can be attributed to low molecular weight additives used for TPE curing. The results of WAXS investigation of the phase structure of TPEs show that in comparison to individual LDPE-pc the introduction of BR into LDPE-pc matrix leads to changes in the angular positions of the WAXS diffraction peaks of LDPE-pc component: the peaks with angular positions 2Θ=21.1ο and 2Θ=23.4o move to the angular positions 2Θ=21.6ο and 2Θ=23.9o, respectively. The positions of the diffraction peaks do not change at introduction and further increase of content of reactive couple in modified TPEs. The corresponding calculation have shown that the mean size of microcrystals <D>=11.4-11.5 nm, and the crystal lattice spacing <d>=0.411. One can see some increase of mean size of LDPE-pc component microcrystals and decrease of crystal lattice spacing in TPE samples (see Table 7) in comparison to pure LDPE-pc (see Table 3). As it was mentioned above the angular positions of diffraction peaks are constant for all TPEs studied, but some change of intensities of the mentioned


5 10 15 20 25 30 35 400

20

40

19.5o

23.9o

21.6o

4

3

2

1I /

1000

pul

ses

2Θ, degree Figure 11. Experimental WAXS curves for unmodified LDPE-pc / BR TPE (1) and LDPE-pc (PE-co-AA)/BR (PB-NCO) with = 1.5 wt.% (2), 7.5 wt.% (3) and 10 wt.% (4) of PB-NCO in BR phase. The ratio of PB-NCO/PE-co-AA was kept 1/1 e.e.w. Beginning from the second curve from the bottom, each next curve was shifted upwards by 5 digits. peaks is observed. This fact has to be reflected in the change of degree of crystallinity <X>, the data are summarized in Table 7. The value of <X> represents the overall crystallinity of blend material and can be compared with the theoretical (additive) value, <X>add, calculated by assuming both retaining by LDPE-pc component its original value of <X>=27.5 %, and the additivity of components contribution. The results show, that the experimental <X> is higher than <X>add, that means, that the rubber component changes the polyethylene crystallization conditions and that its introduction in crystallizable LDPE-pc matrix promotes the phase separation between the crystalline (polyethylene) and amorphous (polyethylene/rubber) phases in TPEs formed. It can be seen that the unmodified TPE has the highest value of <X>. The downtrend of <X> values observed at the introduction of reactive couple in TPEs is an evidence of the destroying some part of crystallites (obviously defected) due to their involving into amorphous phase. This has to be reflected in decrease of onset of melting temperature of crystallites and depression of Tm as it will be shown below by DSC data. In such a case the above downtrend of <X> value can be attributed to the decrease of phase separation of components in modified TPEs. It can be concluded that the experimental results differ from theoretical (additive) data as a result of interactions between the phases, at that the each component affects the microphase structure of another, in TPEs. As a general remark, this fact indicates partial reactively induced compatibilization of BR


and LDPE-pc in TPEs studied. However, the mentioned distinctions are not so significant, indicating the existence of regions with the structure of individual components in all TPEs. TPEs modified by PB-NCO/PE-co-AA are characterized by higher compatibility of components in comparison to unmodified TPE and the optimal content of the PB-NCO/PE-co-AA modifier corresponds to 7.5 % PB-NCO per BR. These results will be confirmed by DSC and DMTA data below.

Table 7. Degree of crystallinity <X> for TPEs produced.

WAXS DSC Composition, wt.%

X, % Xadda), % X,%

LDPE-pc / BR = 60 / 40 (unmodified) b) 19.3 16.5 28 LDPE-pc (PE-co-AA) / BR (PB-NCO) c) : PB-NCO = 1.5 18.5 16.4 26

PB-NCO = 7.5 18.0 16.3 17 PB-NCO = 10.0 18.7 16.2 19

a) <X>add – theoretical (additive) degree of crystallinity: <X >add = X(R-LDPE) · wi , where wi is polyethylene fraction in TPEs; b) LDPE-pc/BR = 60 / 40 ratio is kept for all samples studied; c) NCO / COOH ratio was kept equal 1/1.

5.2. Differential scanning calorimetry Table 7 shows the experimental values of crystallinity degree calculated form DSC data. It is clearly visible the same tendency of <X> value changing and the results obtained agree rather well with WAXS data. Some differences in absolute values have been argued that the two techniques are not directly comparable. Figure 12 shows the Cp (T) plots of TPEs studied. Some depression of melting peak Tm of LDPE-pc component is observed for all TPEs, the summarized data are presented in Table 8. The mentioned melting peak can be related to the melting a crystallizable long polyethylene sequences having a low number of chain defects (branching, graftings, etc.) [32]. Thus, the depression of melting peak in TPEs studied occurs due to increasing defected crystallites content, here mainly because of grafting amide bridges in thermoplastic/rubber interface that has to hamper the crystallization process. One can see that the highest value of depression of melting peak is observed for TPE with content of reactive couple equal 7.5%.


200 250 300 350 4001,6

2,0

2,4

2,8

3,2

Temperature, K

C

p, Jg-1

K-1

388 K

Figure 12. Temperature dependence of specific heat capacity (Cp) of TPEs based on: LDPE-pc/BR (■); LDPE-pc (PE-co-AA)/BR (PB-NCO) TDVs with 1.5 wt.% (□), 7.5 wt.% (○) and 10 wt.% (∇) of PB-NCO in BR phase. The ratio of PB-NCO/PE-co-AA was kept 1/1 e.e.w. 5.3. Dynamic mechanical thermal analysis The DMTA data can give more information about relaxation processes in mixed rubber/polyethylene amorphous phase of TPEs studied. The temperature dependence of storage modulus, E´, loss modulus, E´´, and tangent delta, tan δ are shown in Figure 13 (a, b and c, respectively). In general, one can see that significant differences are observed in relaxation behaviour of LDPE-pc, BR and TPEs based on them, as well as between modified and unmodified TPEs with the same LDPE-pc/BR ratio. One can see that despite of the fact that all TPEs have significant part of crosslinked chains (see Figure 9), they exhibit thermoplastic properties. The character of E´ (T) plots for TPEs at high temperature (>480 K) is typical for thermoplastic polymers and similar to LDPE-pc (see Figure 13, a). We suppose that in all TPEs studied LDPE-pc forms continues phase (matrix) and BR forms a disperse phase. The analyses of temperature dependences of E´´ (Figure 13 b) evidence that even virgin BR and LDPE-pc have two-phase morphology structures. LDPE-pc is characterized by presence of two main transitions: α-transition (Tg1) centred at 333 K is attributed to the relaxation processes in crystalline phase, as mentioned above due to melting and further recrystallization of defected crystallites, and the β-relaxation (Tg2) centred at 243 K is attributed to the relaxation processes of branched chains of amorphous phase of LDPE


200 250 300 350 400 450

1

10

100

200 250 300 350 400 4500

9

18

0

2

4

6

8

200 250 300 350 400 4500,0

0,1

0,2a)

BR

LDPE-pc

E",

MPa

E',

MPa

b)

BR

LDPE-pc

α

β α

Temperature, KTemperature, K

c)

LDPE-pc

BRtan

δ

Temperature, K Figure 13. Temperature dependence of (a) storage modulus, E´, (b) loss modulus, E´´, and (c) tangent delta (tan δ) for LDPE-pc, BR and TPEs based on: LDPE-pc/BR (■);LDPE-pc (PE-co-AA)/BR (PB-NCO) TPEs with 1.5 wt.% (□), 7.5 wt.% (○) and 10 wt.% (∇) of PB-NCO in BR phase. The ratio of PB-NCO/PE-co-AA was kept 1/1 e.e.w. [33]. Note, that high temperature shoulder (centred at ~280 K) observed on β-relaxation transition can be attributed to the presence of crosslinked chains in amorphous phase of LDPE-pc. BR cured is characterized by presence of two main transitions too: the α-relaxation (Tg) centred at 253 K is a relaxation of flexible chains of BR and second high-temperature relaxation with Tonset~300 K (see Figure 13 c) is a relaxation of BR segments limited by intermolecular cross-linking. The characters of E″ (T) and tan δ (T) plots evidence that all TPEs studied are characterized by microphase separation of components and have complicated multiphase structure. This conclusion is confirmed by presence of some overlapped transitions observed in the mentioned plots: a low-temperature transition in the region ~210-300 K that is a result of a superposition of a strong α-relaxation of BR and the weak β-relaxation of LDPE-pc, as well as a high-temperature transition in the region of 320-400 K that is a result of superposition of weak relaxation of BR segments limited by intermolecular cross-linking and the strong α-relaxation of LDPE-pc. At the temperature above ~380 K the melting of crystallizable long polyethylene sequences having a low number of chain defects is started. The temperature positions of α-relaxation peaks taken from corresponding peaks of E″ (T) plot (see Figure 13 b) of the BR-rich and LDPE-pc rich phases in TPEs studied are shown in Table 8. It can be clearly seen that α-relaxation peaks of BR and LDPE-pc are shifted toward one another in modified TPEs in comparison with unmodified TPEs or individual components.


Table 8. Phase transition temperatures for LDPE-pc, BR and TPEs studied.

DSC DMTA α-relaxation peak temperaturea), K

Composition, wt.% Tm, K (for

LDPE-pc)

Onset of Tm, K

BR-rich phase (Tg)

LDPE-pc-rich

phase (Tg1)

LDPE-pc 389 322 - 353 BR - - 253 -

LDPE-pc/BR = 60 / 40 (unmodified)b)

388 246 258 333

LDPE-pc (PE-co-AA)/BR (PB-NCO)c) :

PB-NCO = 1.5 387 243 266 333 PB-NCO = 7.5 385 220 263 328

PB-NCO = 10.0 387 250 265 337 a) The value has been taken from the E’’ peak; b) LDPE-pc/BR = 60 / 40 ratio is kept for all samples studied; c) NCO / COOH ratio was kept equal 1/1. This fact can be explained by the interaction between BR and LDPE-pc phases due to the formation of the essential interface layer mainly based on PB-NCO/PE-co-AA grafting from rubber and polyethylene phases, respectively. In addition, the relaxation processes in amorphous phases is hampered by presence of LDPE-pc crystallites. However, from the data presented in Figure 13 c one can see that some increase of intensity of relaxation transitions in the temperature region 230-295 K is observed for modified TPEs in comparison to unmodified TPE or pure LDPE-pc. This fact can be attributed to the increasing chain mobility in amorphous phases obviously due to increasing unsoundness of crystalline phase of LDPE-pc. These results are agreed to above WAXS and DSC data. In conclusion, DMTA shows that the TPEs can be considered as multiphase systems having at least one crystalline, two amorphous phases of individual components and regions of mixed compositions. We suppose further that LDPE-pc crystalline phase consists of microphases formed by ‘perfect’ crystallites and by ‘defected’ crystallites, while LDPE-pc amorphous phase consists of microphases formed by crosslinked chains and by branched chains. BR amorphous phase consists of microphase formed by cross-linked segments and of one formed by flexible linear BR chains. The mixed microphase consists of both the components grafted by reactive compatibilizers. Thus, the final properties of TPEs are determined by the heterogeneity of the individual components, as well as by the heterogeneity caused by the thermodynamic


immiscibility of the components. The degree of compatibilization is determined and changed, to a large extent, by the grafting reaction of PB-NCO/PE-co-AA reactive compatibilizer and by the formation of the extensive interfacial layer that leads to improving interfacial adhesion between rubber and polyethylene phases. Thermoplastic elastomers based on recycled polyethylene (LDPE or HDPE) and fresh rubber (BR or styrene-butadiene rubber (SBR)) were prepared by using technologies of dynamic vulcanization and reactive compatibilization or plasticization [16, 34]. Structure-property relationships for TPEs prepared have been investigated and effectiveness of different compatibilizers to promote an interfacial adhesion between the components have been compared [16, 34]. Based on these studies the following conclusions can be drawn: For LDPE-pc/BR based TPEs the highest effectiveness of PE-co-AA/PB-NCO reactive couple was fixed whereas for HDPE-pc/SBR based TPEs the most effective reactive couple was PE-co-GMA (glycidyl methacrylate)/PB-NH2 (amino-functionalized polybutadiene). The values of TS and EB increased respectively by 31% and 63% for LDPE-pc (PE-co-AA) / BR(PB-NCO) TPE, and by 87 % and 182 % for HDPE-pc (PE-co-GMA)/SBR (PB-NH2) TPE. The investigation of phase structure has shown that some destroying and increasing mean size of crystallites, as well as their involving into amorphous phase are observed due to decreasing phase separation of components in all the modified TPEs. Growth of Tg value has been fixed for all the modified TPEs in contrary to unmodified ones or individual components that is a result of formation of mixed polyolefin/rubber amorphous phase with the essential interface layer consisted of both components grafted by reactive compatibilizers used. 6. Thermoplastic elastomers based on recycled HDPE, EPDM and reclaimed ground tire rubber High performance TPEs based on recycled HDPE (or LDPE, or PP), olefinic type ethylene/propylene/diene monomers (EPDM) containing rubber and GTR treated with bitumen were prepared by using the dynamic vulcanization technology, and their structure-property relationships have been investigated. It was established that at special pre-treatment of GTR by bitumen bestows on the resulting TPEs outstanding mechanical properties [17, 28, 29, 30, 34-36]. The method of GTR reclamation used is described in Section 3 of this Chapter. The recipes and processing conditions used for the TPE compositions are given in Table 9 and Table 10, respectively. In one test series (cf. Table 10) a mastication of composition was carried out in the kneading chamber of a


Brabender plasticorder (model PL 2000) at 160 oC and 80 rpm. Recycled HDPE-pc was melted first for 2 min, then EPDM was added and melted for 2 min, and finally the GTR or GTR/bitumen blend (1/1 by weight) was added and masticated with other components for a further ~10 min.

Table 9. Composition recipes used.

Component content, wt.% Composition Recipe “a” Recipe “b” Recipe “c”

HDPE-pc/rubbera) 53.3/46.7b) 61.5/38.5c) 50/50 HDPE-pc/rubbera)/GTR 40/35/25 40/25/35 50/25/25

HDPE-pc/rubbera) / (GTR/bitumen)

40/35/25(1/1) 40/25/35(1/1) 50/25/25(1/1)

a) EPDM was used as a rubber in the blends of the recipes “a” and “b”; SBR was used as a rubber in the blends of the recipe “c”. b) Ratio of HDPE-pc/rubber = 53.3/46.7 (wt.%) is equal to 40/35 (wt.%) in other blends of the recipe “a”. c) Ratio of HDPE-pc/rubber = 61.5/38.5 (wt.%) is equal to 40/25 (wt.%) in other blends of the recipe “b”.

Table 10. Conditions and codes of producing methods used. Conditions of producing method Code

Mastication by using Brabender plasticorder 1. Mastication of composition (in Brabender plasticorder). A 1. Heating of GTR/bitumen blend. 2. Mastication of HDPE-pc/rubber/(GTR/bitumen) blend followed by its rolling.

B

1. Heating of GTR/bitumen blend followed by its rolling. 2. Mastication of HDPE-pc/rubber/(GTR/bitumen) blend.

C

1. Heating of GTR/bitumen blend followed by its rolling. 2. Mastication of HDPE-pc/rubber/(GTR/bitumen) blend followed by its rolling.

D

Mastication by using single-screw extruder

1. Mastication of GTR/bitumen blend followed by its granulation. 2. Mastication of HDPE-pc/EPDM/(GTR/bitumen) blend.

E

1. Mastication of GTR/bitumen blend followed by its rolling and granulation. 2. Mastication of HDPE-pc/EPDM/(GTR/bitumen) blend.

F

1. Heating of GTR/bitumen blend followed by its mastication in extruder and granulation. 2. Mastication of HDPE-pc/EPDM/(GTR/bitumen) blend.

G

1. Heating of GTR/bitumen blend followed by its rolling, mastication in extruder and granulation. 2. Mastication of HDPE-pc/EPDM/(GTR/bitumen) blend.

H


In the other test series (cf. Table 10) a mastication of compositions was carried out by using one-screw extruder (model PLV 150). First, the GRT powder was extruded together with the bitumen at the temperature profile of 155/165/175 °C and the screw speed 40 rpm followed by granulation of the extrudate. Afterwards, the granulated GTR/bitumen blend was extruded again together with other components (T=155/165/175 °C, 40 rpm). Some compositions (cf. Table 10) were prepared by rolling. GTR/bitumen blends and the related TPE compositions were produced on mill rolls at T~60 °C for 40 min. 6.1. Tensile properties Tensile properties of the GTR containing compositions produced by different mastication methods (cf. Table 9) are shown in Table 11. The experimental data clearly demonstrate the beneficial effect of bituminous treatment of GTR and its dependence on the mastication method chosen [28]. One can see that GTR-containing TPEs with suitable tensile properties can be obtained for all recipes used by choosing proper production methods. Based on above-mentioned definition of TPEs (cf. section “Introduction”), one can conclude that the samples B3-B6, B10-11, B14 and B15-B18 satisfy to qualifying standards for TPEs. Note that compression set values are not reported here, however, the related values were below 50 %. It is clearly seen that the requisite condition for producing GTR containing TPEs with suitable properties is the preheating of GTR/bitumen blend before the mastication with the other blend components in Brabender plasticorder or extruder. Sol-gel analysis has shown that heating treatment of GTR leads to decreasing gel-fraction content by ~ 8 % (compared to initial GTR), that is a result of partial devulcanization of GTR. Further decrease of the gel-fraction content up to ~13 % is observed for GTR/bitumen blend after its preheating (calculation has been done per GTR content). It can be concluded that in such a case bitumen acts as a softening and devulcanizing agent for GTR breaking-up sulfuric crosslinks in GTR and therefore leading to activation and functionalization of at least its surface. The reactive sulfur released from GTR and sulfur of bitumen components (cf. Table 6) further take part in covulcanization of pre-heated GTR/bitumen blend with a fresh rubber (EPDM) in the revulcanization step. Indeed it can be seen that the tensile strength and ultimate elongation are higher for all TPEs produced by methods included the procedure of GTR/bitumen pretreatment compared to those produced by the other methods. This suggests an effective interfacial stress transfer between the matrix and the GTR particles [14] due to a better entanglement of the partly devulcanized GTR rubber chains into surrounding matrix.


One can see that the higher GTR content in TPEs the lower tensile properties of the product. Additional homogenization by rolling pre-heated GTR/bitumen blend and/or final product (cf. Table 11) further improved the tensile strength and, especially, ultimate elongation of the resulting TPEs. This fact evidences of an effective compatibilization of the TPE blend components [37]. Table 11. Tensile properties of individual polymers and blends produced. The data presented in the parentheses represent the properties after ageing at 70 oC for 24 hours.

Blend code

Composition Code of producing

method

Tensile strength,

MPa

Ultimate elongation, %

Hardness (Shore A)

Mastication by using Brabender plasticorder Recipe “a”

B1 HDPE-pc/EPDM A 13.0 (20.0) 840 (754) 93 B2 HDPE-pc/EPDM/GTR A 4.4 (6.7) 114 (120) 96 B3 HDPE-pc/EPDM/ (GTR/bitumen) A 4.1 (6.5) 168 (176) 94 B4 HDPE-pc/EPDM/

(GTR/bitumen) B 4.9 (8.0) 528 (536) 93

B5 HDPE-pc/EPDM/ (GTR/bitumen)

C 6.0 (10.2) 540 (535) 93


D 6.1 (10.7) 615 (590) 93

Recipe “b” B7 HDPE-pc/EPDM A 11.6 (17.9) 750 (593) 90 B8 HDPE-pc/EPDM/ GTR A 3.7 (6.7) 46 (58) 95 B9 HDPE-pc/EPDM/ (GTR/bitumen) A 3.9 (7.0) 85 (97) 93


B 4.0 (5.2) 127 (194) 93


C 5.9 (5.3) 377 (325) 92

Recipe “c” B12 HDPE-pc/SBRa) A 6.8 265 83 B13 HDPE-pc/SBRa)/GTR A 7.0 18 79 B14 HDPE-pc/SBRa)/(GTR/bitumen) C 8.4 355 84

Mastication by using single-screw extruder Recipe “a”

B15 HDPE-pc /EPDM/ (GTR/bitumen)

E 7.5 270 88


F 9.0 300 89


G 9.8 425 86


H 13.6 515 86

a) Curing system used, phr per 100 phr of SBR in mix formulations: Sulfur 0.9; Tetramethyl thiuram disulfide 0.1 0.1; ZnO 5.0; Stearic acid 2.0; 2-mercaptobenzothiazole 1.0.


To check the compatibilization efficiency of bitumen in rubber/polyolefin blends, TPEs containing SBR were selected (samples B15-B17). Note that SBR is fare less compatible with HDPE than EPDM. It can be seen that tensile characteristics of HDPE-pc/SBR TPE (sample B12) is much lower compared to HDPE-pc/EPDM TPE (sample B1). Introduction of GTR into HDPE-pc/SBR formulation (sample B13) leads to a dramatic reduction in ultimate elongation compared to HDPE-pc/EPDM/GTR blend (sample B2). However, after bituminous treatment of GTR and further melt production of HDPE-pc/SBR/(GTR/bitumen) TPE (sample B14) the tensile properties were strongly improved compared to the reference HDPE-pc/SBR blend (sample B13) is observed. Hence it can be concluded that bitumen acts as effective compatibilizer for polyolefins and rubbers in TPEs [28]. Note that all EPDM-based compositions exhibit almost similar hardness values, which are near or above 90 Shore A units, whereas the SBR-based compositions exhibit lower hardness values, which are above 80 Shore A. Table 11 indicates some increase of tensile properties for the most of the compositions studied as a result of thermal ageing. It can be concluded here that all the blends studied were found to be remarkably stable to ageing. As it was shown in [14] the enhancement in tensile strength and marginal change in ultimate elongation suggest the formation of some additional crosslinks (post-curing). On the other hand, the increasing both the tensile strength and the ultimate elongation observed for the samples B2-B5 and B8-B10 suggest that a post-curing occurs mainly in rubber phase. TPEs based on LDPE-pc (or HDPE-pc), EPDM and GTR (or GTR/bitumen) were investigated using TMA techniques [38]. It is shown that the presence of GTR/bitumen component leads to decreasing value of thermal expansion coefficient of the resulting TPEs due to diffusion of bitumen into thermoplastic matrix. The compositions of TPEs studied were optimized. 6.2. Thermogravimetric analysis (TGA) The thermal stability of GTR containing compositions (sample B2) and GTR/bitumen containing (sample B6) was studied by TGA and compared to that of the reference HDPE-pc/EPDM TPE (sample B1) [17, 28, 39], the corresponding curves are shown in Figure 14. It can be seen that an introduction of GTR into the basic HDPE-pc/EPDM blend is not accompanied by significant changes in thermal stability. Some shift to higher temperatures is observed for the stages of intensive decompositions above ~500 oC only, however, finally a char residue values are near the same for the both samples: 1.77 % for the reference


HDPE-pc/EPDM (sample B1), and 1.67 % for the HDPE-pc/EPDM/GTR (sample B2). In addition, one can see that the stage of thermal oxidative destruction at 150-250 oC characteristic for the basic HDPE-pc/EPDM blend disappears completely for the GTR containing samples. The thermal behavior of GTR/bitumen containing TPE (sample B6) in the temperature region below ~340 oC is quite similar to the GTR containing blend (sample B2). However, some depression by ~28-33 oC in the temperature of the maximal rate of decomposition in the region from ~340 oC to ~460 oC was observed for the sample B6 compared to the both other samples (B1 and B2), in the temperature region above ~460 oC the thermal behavior of the sample B6 is quite similar to the reference B1. Similar results were obtained [17] for analogous TPEs based on LDPE-pc and PP-pc.

0 200 400 600

Δm

, %

HDPE-

-pc:EPDM HDPE-pc:EPDM:GTR HDPE-pc:EPDM:GTR/bitumen

exo

Δm

τ-1, %

min

-1

0 200 400 600-1,5

-1,0

-0,5

0,0

0 200 400 600

-100

-50

0

Temperature, oC Figure 14. Thermogravimetric analysis of the blendes: (─) HDPE-pc/EPDM (sample B1); (○) HDPE-pc/EPDM/GTR (sample B2), and (▲) HDPE-pc/EPDM/(GTR/bitumen) (sample B6).


6.3. Rheological properties The rheological properties of the HDPE-pc/EPDM/GTR (sample B2) and HDPE-pc/EPDM/(GTR/bitumen) (sample B6) were studied and compared with those of the reference HDPE-pc/EPDM (sample B1) [17, 28]. The dependence of shear viscosity (η) versus shear rate (γ) obtained at different temperatures is shown in Figure 15. First of all it should be noted that the measurements were not possible with HDPE-pc/EPDM/GTR blends because of their very high viscosity (>105 Pa·s) caused by the introduction of crosslinked GTR particles. A significant decrease in the melt viscosity was observed for GTR/bitumen containing TPE (sample B6) compared to the reference HDPE-pc/EPDM (sample B1) at a given temperature and a shear rate (except the region of the highest shear rate). Undoubtedly, this is mainly due to the low molecular weight of bitumen which acts as an effective plasticizer in the TPE studied. For both samples B1 and B6 the viscosity decreases with increasing shear rate at a fixed temperature. Shear thinning is typical for most thermoplastic polymers [8]. However, a viscosity increase was observed for sample B1 with rising temperature in the low shear-stress region [28]. On the other hand, in the high shear stress region the opposite tendency is obvious. The above mentioned increase in viscosity can be explained by post-curing of the TPE (sample B1) during the rheological measurement (i.e. thermally induced

10-2 10-1 100 101103

104

105

180 oC 190 oC 200 oC

180 oC 190 oC 200 oC

She

ar v

isco

sity

, log

η,

Pa ⋅

s

Shear rate, log γ , s-1 Figure 15. Shear-viscosity (η) vs. shear-rate (γ) plots for HDPE-pc/EPDM (sample B1, open symbols) and for HDPE-pc/EPDM/(GTR/bitumen) TPE (sample B6, solid symbols) obtained at different temperatures (indicated in the plot). The codes of the samples correspond to the blends in Table 11.


crosslinking of the EPDM). It is intuitive that the higher temperature, the higher the crosslinking of the product [40]. Note that the gel fraction of sample B1 increased from zero (before mastication) up to ~16 % (after mastication). Note that a remarkable influence of the rubber component on the flow behavior of blends exactly in the range of low shear-stress region occurs, where relaxation processes take place [41] and some agglomerated structures can be formed [42, 43]. Logically, in the high shear-stress region the influence of rubber component on flow behavior of the TPE is insignificant due to the absence of relaxation processes and destruction of agglomerated structures [40]. Very similar results were obtained for TPEs based on LDPE-pc and PP-pc [17]. 6.4. Scanning electronic microscopy (SEM) Figure 16 depicts SEM photomicrographs taken from the cryogenic fracture surfaces of the sheets of some blends. They were produced according to recipes “a” (samples B2, B3 and B4) and “b” (samples B8, B9 and B11) via different methods. The codes of the samples correspond to the blends in Table 11. One can clearly seen that GTR particles directly dispersed in HDPE-pc/EPDM blend (samples B2 and B8) are very poorly bonded to the matrix, a lot of large and small size GTR particles are observed outside the matrix indicating for lacking interaction between them [17]. The sample B8 produced with the higher GTR content (35 wt. %) is characterized by deteriorated homogeneity of the surface compared to the samples B2, i.e. increase of an apparent size of de-bonded GTR particles and a presence of cracks (or holes) are observed. Both samples exhibit unacceptable low tensile properties (cf. Table 11), especially, the sample B8 [28]. It can be seen that the surface of the HDPE-pc/EPDM/(GTR/bitumen) blends produced by mastication of compositions by Brabender plasticorder (samples B3 and B9, method “A”) look more homogeneous. Furthermore, the apparent size of the GTR particles is reduced, the small GTR particles are well incorporated into the matrix, whereas the larger ones are partially protruded outside the fracture surface, and there are no visible holes or cracks compared to the corresponding bitumen-free samples. It can be concluded that in such a case bonding between GTR particles and thermoplastic matrix is improved that results into some increasing elongation at break (cf. Table 11). However, the homogeneity level achieved does not provide the high level of tensile properties for samples B3 and B9 (cf. Table 11). We consider that no significant interfacial layer between GTR particles and thermoplastic matrix is formed.


Figure 16. Typical SEM photomicrographs of cryo fractures cut surfaces of TPEs of recipes “a” (samples B2, B3 and B4) and “b” (samples B8, B9 and B11). The codes of the samples correspond to the blends in Table 11. A better bonding between GTR particles and matrix have been reached for the samples B4 and B11 produced by the methods “B” and “C”, correspondingly (cf. Table 10), where the partial devulcanization of GTR under the preheating of GTR/bitumen blend is happened. The surface of the samples B4 and B11 is characterized by higher level of homogeneity compared to the other samples. It can be considered that it is a result of formation of a significant interfacial layer of partially devulcanized GTR, bitumen and other components of the blends. Understandably, the samples B4 and B11exhibit high tensile properties (cf. Table 11), especially it is relative to the sample B4.


6.5. Differential scanning calorimetry (DSC) Typical DSC curves for the individual polymers and for TPEs produced are shown in Figure 17 a and b, respectively, and the corresponding thermal characteristics are summarized in Table 12. Both components (EPDM and HDPE-pc) keep their own amorphous and crystalline phases in TPEs produced, however some reduction in the crystallinity (Xc) can be noticed, especially for the EPDM component [17, 28]. The dramatic reduction in the Xc values of the EPDM component can be explained that its crystallization is hampered due to the presence of HDPE-pc crystallites and intermingling of several EPDM chains with those of the HDPE-pc. The outcome of the latter is a “mixed amorphous phase” for HDPE-pc/EPDM [17, 28, 39]. Some depression of the melting temperature (Tm) values of both EPDM and HDPE-pc components of the TPEs compared to the individual polymers was observed. This depended on the composition and processing conditions used. It is known that the depression of Tm of polymers in the blends is caused by the formation of less perfect crystallites or crystallites having a smaller size [44]. Irrespective which one is at work Tm decrease is always evidence for improved blend compatibility.

-100 -50 0 50 100 150 200

EN

DO

EN

DO

HDPE-pc

EPDM

bitumen

a)

-100 -50 0 50 100 150 200

B10

B9

B6

B3 B2

B1

b)

Temperature, oC

Figure 17. Typical DSC traces for (a) individual HDPE-pc, EPDM and bitumen and for (b) TPEs produced by different methods. The codes of the curves correspond to the formulation codes in Table 11.


Table 12. DSC characteristics for individual components and blends produced.

Tm, oC Tm onset / Tm end, oC Xca), % Blend

code Composition

EPDM HDPE-pc EPDM HDPE-pc EPDM HDPE-pc

- HDPE-pc - 136 - 37 / 160 - 65 - EPDM 47 - 28 / 65 - 12 - Recipe “a”

B1 HDPE-pc/EPDM 45 135 26 / 69 73 / 156 6 57 B2 HDPE-pc/EPDM/GTR 46 131 26 / 57 73 / 145 4 63 B3 HDPE-pc/EPDM/(GTR/bitumen) 45 130 27 / 65 73 / 143 3 61 B4 HDPE-pc/EPDM/(GTR/bitumen) 44 129 29 / 58 73 / 143 3 56 B6 HDPE-pc/EPDM/(GTR/bitumen) 44 132 29 / 65 70 / 147 4 62

Recipe “b” B7 HDPE-pc/EPDM 45 135 27 / 70 73 / 158 4 60 B8 HDPE-pc/EPDM/GTR 45 132 36 / 57 70 / 146 3 65 B9 HDPE-pc/EPDM/(GTR/bitumen) 42 130 29 / 64 70 / 142 2 64

B10 HDPE-pc/EPDM/(GTR/bitumen) 42 131 28 / 70 73 / 144 2 50 B11 HDPE-pc/EPDM/(GTR/bitumen) 45 133 32 / 58 75 / 146 0 60

a) The Xc (crystallinity) values were calculated taking into account the weight fraction of PE in the EPDM (~71 %) and that of EPDM in the blends; the enthalpy of melting of PE with 100 % degree of crystallinity was taken as 283 J/g. In addition, for HDPE-pc matrix a significant shift of onset of melting temperature (Tmonset) towards higher temperature and narrowing of a region of crystallites melting (Tm end-Tmonset) is observed in all the blends (cf. Table 12) compared to the individual HDPE-pc. The growth of Tm onset can be caused by a disappearance of smaller the less perfect crystallites due to their involving into the amorphous phase and the narrowing of region of crystallites melting is a result of decreasing crystallite dimension dispersion. 6.6. Dynamic mechanical thermal analysis (DMTA) Temperature dependencies of loss modulus (E"), storage modulus (E′), and loss factor (tan δ) for individual HDPE-pc and EPDM as well as for some blends produced are shown in Figures 18, 19 and 20, respectively [28]. The glass transition temperature (Tg) values defined as a temperature position of E" peak, and corresponding E" values at Tg’s are listed in Table 13. EPDM has one sharp relaxation peak at –36 oC (α-transition), corresponding to the Tg of its amorphous phase (cf. Figure 18). Some rise in E′ and tan δ values around ~50 oC (cf. Figures 19 and 20) can be attributed to the melting of residual polyethylene crystallites that is confirmed by the above DSC data. HDPE-pc has two broad relaxation peaks (cf. Figure 18) at –15 oC (β-transition) assigned to the Tg of its amorphous phase, and the peak around ~60 oC (αc-transition) relating to the vibration and rotational motion of –CH2–


groups in crystalline phase due to recrystallization of smaller the less perfect crystallites [45]. For both EPDM and HDPE-pc the low temperature transition below -100 oC is observed due to the crankshaft mechanism of –CH2–CH2– polyethylene chain segments [46]. The character of E' (T) plots (cf. Figure 19) of all the blends studied is quite similar to the HDPE-pc that means that HDPE-pc forms a continuous thermoplastic phase (matrix), while the dispersed phase is formed by the EPDM/GTR mixture.

Figure 18. Temperature dependence of loss modulus (E") for HDPE-pc, EPDM and TPEs produced. The codes of the curves correspond to the formulation codes in Table 11.

Figure 19. Temperature dependence of storage modulus (E′) of individual HDPE-pc, EPDM and TPEs produced. The codes of the curves correspond to the formulation codes in Table 11.


-100 -50 0 50 100 1500,0

0,1

0,2

0,3

0,4

0,5EPDM

HDPE-pc

B7 B8 B9 B11

Temperature, oC

tanδ

Figure 20. Temperature dependence of loss factor (tan δ) for HDPE-pc, EPDM and TPEs produced. The codes of the curves correspond to the formulation codes in Table 11. The introduction of GTR into HDPE-pc/EPDM TPE (sample B8) results in essential changing in its viscoelastic properties. New sharp relaxation transition around -50 oC (Tg1) appear (cf. Figures 18 and 20) that is characteristic for rubber component of GTR [21]. A considerable lowering by 9÷14 oC in both other Tg’s (cf. Table 13), as well as some downtrend in the E'=f(T) curve and uptrend in the tan δ=f(T) and E"=f(T) curves were observed compared to the GTR-free sample B7 (cf. Figures 18-20). All these changes suggest a significant growth of chain flexibility of the components of the GTR-containing sample B8 due to disordering the thermoplastic matrix by dispersed crosslinked GTR particles caused by poor interphase adhesion between the components [18, 19]. The tensile characteristics of the sample B8 are much lower compared to the GTR-free sample B7 (cf. Table 11). Introduction of bitumen into HDPER/EPDM/GTR blend (sample B9) yields a convergence between Tg2 and Tg4 values (cf. Table 13) [17, 28] and growth by ~80 % ultimate elongation value compared to the sample B8 (cf. Table 11) that can be interpreted as improved “mixing” of the blend components. However, the disordering thermoplastic matrix by dispersed rubber particles keeps sufficiently high that is reflected by the high values of E" and tan δ (cf. Figures 18 and 20, respectively). As a result, the tensile properties of the sample B9 (cf. Table 11) do not satisfy to qualifying standards for TPEs [37]. It can be concluded that in such a case the processing conditions used (method “A”) do not provide the required devulcanization degree of GTR and interface adhesion between the components. Some growth in the crosslink degree of the amorphous phase of the blend takes place, which is confirmed by both the downtrend of E"=f(T) curve in the temperature region below ~-50 oC and the uptrend of the E'=f(T) curve (cf. Figures 18 and 20, respectively). We consider this to be a result of dynamic vulcanization of dispersed rubber


phase inside the plastic matrix. So, the bitumen can act as an additional curing agent. This is the reason why the term dynamic vulcanization can be used although bitumen is not at all a traditional curative for rubbers. The HDPE-pc/EPDM/(GTR/bitumen) blend (sample B11) produced by the method “C” is characterized by significant reduction of E" and tan δ values (cf. Figures 18 and 20) as well as a further uptrend of the E'=f(T) curve (cf. Figure 19) compared to sample B9. Undoubtedly, such changes evidence a further growth in the crosslink degree of the dispersed EPDM/GTR rubber phase in sample B11 compared to B9. However, based on the increasing ultimate elongation value of sample B11 (cf. Table 11) an improved interfacial adhesion between the components can be unequivocally quoted. We consider that this is due to the higher degree of bitumen induced devulcanization of GTR at preheating of GTR/bitumen blend before mastication additionally reflected by the significant decrease of Tg1 onset of rubber phase (cf. Figure 18) from –65 oC (for the sample B9) to –75 oC (for the sample B11). This result agrees well with the conclusions made on the base of sol-gel analysis (cf. section “Tensile Properties”). As it was noted above the (re)covulcanization of partly devulcanized GTR with EPDM and bitumen is occurred during mastication of composition in Brabender plasticorder and finally, sample B11 exhibits high values of tensile strength and ultimate elongation (cf. Table 11). Certainly, this is a result of the improved compatibility of the blend components. Based on analysis of the DMTA data it can be concluded that at producing GTR/bitumen-containing TPE by method “C” the bitumen first acts as devulcanizing agent for GTR and then simultaneously as an effective

Table 13. DMTA data for individual components and blends produced.

Tg, oC / E"a), MPa for phases rich in:

GTR EPDM amorphous

HDPE-pc

amorphous HDPE-pc crystalline

Blend code

Composition

Tg1 / E" Tg2 / E" Tg3 / E" Tg4 / E"

- HDPE-pc - - -15 / 65 60 / 100 - EPDM - -36 / 102 −b) - B7 HDPE-pc/EPDM - -31 / 134 −b) 68 / 30 B8 HDPE-pc/EPDM/ GTR -50 / 160 -42 / 150 −b) 54 / 21

B9 HDPE-pc/EPDM/ GTR/ bitumen -49 / 121 -35 / 158 −b) 42 / 35 B11 HDPE-pc/EPDM/ GTR/ bitumen -48 / 63 -33 / 110 −b) 53 / 41

a) E" value taken at Tgi b) Tg3 is overlapped with Tg2.


curing agent for dispersed EPDM/GTR rubber phase and as a compatibilizer for blend components improving the interfacial adhesion between dispersed rubber phase and plastic HDPE-pc [17, 28, 39] or LDPE-pc [39] matrix. 7. Effect of multi-reprocessing on structure and properties of thermoplastic elastomers based on recycled polyolefins and reclaimed ground tire rubber As is known [1], processing of TPEs and preparation of finished products at elevated temperatures and dynamic shear stresses can be accompanied by thermal degradation and crosslinking of polymer chains as well as by the post-vulcanization of rubber through residual unsaturated bonds; as result, all characteristics of TPEs can change. In [47], one can find a detailed description of the mechanisms of the above processes. In the TPEs under study, post-vulcanization can proceed through double bonds of the reclaimed GTR (GTRr), which form during the regeneration of rubber crumbs, unsaturated bonds of EPDM, and free sulfur existing in the system. Crosslinking of the polyolefin matrix can result from the stress-induced degradation of macromolecules and formation of free radicals, which are able to participate in the formation of transversal crosslinks. The above processes primarily affect the flow characteristics of the material. Taking this into account, it seems interesting to study the effect of repeated processing on the rheological characteristics of TPEs under study. Figure 21 presents the typical dependences of effective viscosity (lg η) on shear rate (lg •

γ ) for the samples of initial TPE and TPEs after the first, second, and third processing cycles in the extruder (TPE-1, TPE-3, TPE-6) [48]. As follows from Figure 21, under the selected experimental conditions (T=190 and 210 °C), all TPEs preserve their flowability; in this case, the effective viscosity of all TPEs decreases with increasing temperature (from 190 to 210 °C) and with increasing shear rate. This character of rheological curves seems quite expectable and shows that TPEs preserve their thermoplastic characteristics even after six processing cycles in an extruder. Figure 22 presents the effective viscosity measured at a fixed shear rate ( •

γ = 0.132 s–1) vs. number of processing cycles of TPE samples. As is seen, viscosity of the test samples increases up to the third processing cycle (sample TPE-3); however, starting with the fourth cycle, viscosity decreases and approaches values typical of the initial TPE or even decreases below this value. The corresponding curves at different (fixed) shear rates show similar profiles. As it is known [49], extrusion processing of polymer materials, including polyolefin–rubber


0,01 0,1 1 10

104

•

1' 2' 3' 4'

1 2 3 4

lg

η, P

a⋅s

lg γ , s-1

Figure 21. Viscosity vs. shear rate for the following samples: Initial TPE (1,1'), TPE-1 (2, 2′), TPE-3 (3, 3′) and TPE-6 (4, 4′) at 190 оС (1-4) and 210 оС (1′-4′).

0 1 2 3 4 5 6

15

20

25

N

η, k

Pa⋅s

1

2

Figure 22. Viscosity vs. number of processing cycles N of TPE samples at (1) 190 and (2) 210 °C. TPEs, can be accompanied by two parallel competing processes: crosslinking and thermomechanical degradation of polymer chains of the components. Evidently, crosslinking should increase the melt viscosity, but degradation should decrease it. Therefore, one can conclude that, during the first three processing cycles, processes of post-vulcanization of rubber phase and/or minor crosslinking of the polyolefin matrix play a key role; during further processing of TPE samples (samples TPE-4–TPE-6), thermomechanical degradation processes in polymer chains start to dominate. This conclusion also follows from analysis of the dependences of the activation energy of


flow Ea for the TPE, TPE-3 and TPE-6 samples over the whole range of (Fig. 23). Indeed, one can see that, for the TPE-3 sample, its Ea value is higher than that of the initial TPE; for the TPE-6 sample, this value is lower. Let us assume that this behavior results from the development of crosslinked regions in TPE-3, and these regions hinder the melt flow of TPE; for sample TPE-6, this behavior can be explained by the degradation of polymer chains. Let us mention that, in this case, we consider the intervals of medium and high shear rates because the character of the flow of the polyolefin matrix in TPE is known to be nearly independent of the presence of the crosslinked dispersed rubber phase [41–43]. The point is that, at high shear rates, relaxation processes are almost completely absent [41] and breakdown of agglomerated structures takes place [42, 43]. Figure 24 presents the curves illustrating the content of gel fractions in the TPE samples, and this evidence supports the above conclusions. As follows from Figure 24, the content of the gel fraction in the TPE samples increases as the number of processing cycles is increased to three; later, this content slightly decreases (cycles 4–6). Hence, as compared with the initial TPE, the effective degree of crosslinking increases from the first to the third cycle and decreases from the fourth to the sixth cycle of TPE processing. In this case, the TPE-3 sample is characterized by the maximum degree of crosslinking. Let us mention that the above difference in the rheological characteristics (η, Ea and wg) of the TPE samples is small; hence, under the selected processing conditions, chemical crosslinking and degradation in TPEs are insignificant and exert almost no effect on the characteristics of the TPE-1–TPE-6 samples. For example, as follows from Figure 24, the density of the samples slightly changes, but its tendency to increase with increasing the number of processing cycles is evident. Figure 25 presents the results of mechanical tests before and after repeated processing of the TPE samples in the extruder. As compared with the initial TPE sample, the strength characteristics of the TPE-1–TPE-6 samples are seen to be somewhat improved. In this case, the maximum increase in tensile strength σbr and relative elongation at break br amounts to about 15%. It is worth mentioning that, in the TPE1–TPE-3 samples, the strength remains virtually unchanged but br increases. As it is known [20], increased relative elongation at break in the polyolefin–rubber TPEs suggests better compatibility between components. We can also assume that the above crosslinking processes take place primarily either inside the dispersed rubber (EPDM/GTRr) phase or at the interfacial rubber–polyolefin boundary because, upon crosslinking of thermoplastic polyolefin matrix, br should be decreased. For the TPE-4–TPE-6 samples, one can observe a decrease in br


and a certain increase in σbr. This behavior is likely stem from the development of scarce crosslinks in the polyolefin matrix via the interaction of free radicals, which are formed due to the above degradation processes of PE chains [47]. It seems interesting to study the effect of repeated processing on the thermal stability of the TPE samples. For this purpose, thermogravimetric experiments were conducted [17, 28, 48]. The results obtained suggest that the number of the multi-processing cycles has almost no effect on the thermooxidative degradation of all TPE samples under study, because all curves nearly coincide, and the difference in their thermal characteristics (weight loss, weight loss rate, char residue, etc.) was small.

0,01 0,1 1 1020

25

30

35

40

•

2

3

Å

à, kJ/

mol

lg γ , s-1

1

Figure 23. Activation energy of viscous flow vs. shear rate •

γ for the following samples: (1) initial TPE, (2) TPE-3, and (3) TPE-6.

0,954

0,956

0,958

0,960

0 1 2 3 4 5 60

10

20

30

2

1

wg,

%

ρ, g

/ñm

3

N

Figure 24. (1) Content of gel fraction and (2) density vs. number of processing cycles N of TPE samples.


400

450

500

550

600

0 1 2 3 4 5 66

8

10

12

2

1 ε br,

%

σ br, M

Pà

N

Figure 25. (1) Tensile strength and (2) relative elongation at break vs. number of processing cycles N of TPE samples. Table 14. Characteristics of the crystalline phase (DSC data) of individual components and HDPE matrix of the TPE samples before and after their multi-reprocessing in extruder.

Sample Тm, оС ΔТm, оС ΔНm, J/g Χ, % HDPE-pc 135 63 204 70

Initial TPE 128 58 175 60 TPE-1 128 54 172 59 TPE-2 128 55 170 58 TPE-3 128 51 160 55 TPE-4 128 51 165 57 TPE-5 128 51 150 52

Note: The Hm and X values are calculated with allowance for the weight fraction of the PE component in the system. As was shown earlier in [17, 28, 34, 35, 39], this type of TPE is characterized by a multiphase structure, which involves crystalline and amorphous phases of PE component and mixed amorphous phase, which contains the macromolecules of HDPE, EPDM, and GTRr. It seems interesting to study the effect of the number of processing cycles on the characteristics of crystalline phase of the HDPE matrix in the TPE sample: namely, on changes in its melting temperature Tm and degree of crystallinity X. The relevant data obtained by the DSC method are presented in Table 14. Basic on the DSC results we concluded [48] that these TPEs belong to the


class of semicrystalline polymer blends. As follows from Table 14, the HDPE-pc component in the TPE samples is characterized by lower values of Tm, ΔTm, ΔHm, and X as compared with those of initial HDPE, and this unequivocally suggests the breakdown of some crystallites. The melting temperature of the TPE, TPE-1–TPE-5 samples is the same, but the melting temperature of the TPE-6 sample is 3°C less. At the same time, one can observe an evident tendency toward narrowing of the melting temperature interval and this effect is most pronounced for the TPE-3–TPE-6 samples. A marked decrease in the heat of fusion ΔHm and, correspondingly, a decrease in the degree of crystallinity are also observed for the TPE-3–TPE-6 samples, especially for the TPE-6 sample. All these facts suggest that, after the third processing cycle, a certain part of the most difficult if PE crystallites breaks down and they are transformed into the amorphous phase of TPE, which, as was shown in [28], presents the mixed HDPE-pc–EPDM–GTRr phase. As follows from the literature data [50], for polyolefin–rubber TPEs, the lower Tm and X of the polyethylene component, the higher the compatibility between components in TPEs. Analysis of thermograms of the TPE samples [48] also shows that, after the third processing cycle, the specific heat capacity Cp of the compositions decreases by 0.5–0.6 J/(g K) over the entire temperature interval under study. Usually, the specific capacity Cp of polymer blends decreases with increasing packing density of macrochains of the system components, and this behavior can be explained either by crystallization or chemical crosslinking [51]. As follows from Table 14, with increasing number of processing cycles, the fraction of the crystalline phase in the TPE samples decreases. Taking into account this observation, the decrease in Cp is likely to be related to crosslinking processes. For the TPE-4–TPE-6 samples, a slight increase in Cp can come from the concomitant degradation processes (see the above speculations concerning Rheological characteristics and TGA data). However, the TPE-3–TPE-6 samples are characterized by lower Cp values as compared with those of the initial TPE and TPE-1 and TPE-2 samples, and this observation proves the existence of scarce crosslinks in the HDPE matrix. This evidence agrees fairly well with the physical-mechanical tests (Fig. 25). DMTA studies make it possible to estimate the effect of repeated processing on the characteristics of the amorphous phase of the TPE samples. Figure 26 presents the temperature dependence of mechanical loss tangent for HDPE-pc and EPDM and for the TPE samples. The results of DMTA study of the HDPE-pc, EPDM and their TPEs with GTRr and GTR/bitumen are analyzed in Section 6. Here we can compare the results obtained for these TPEs and for them after multi-reprocessing. The data presented in Figure 26 and Table 14 show that, as compared with the initial TPE, Tg of the TPE-1– TPE-5


-60 0 60 120

0

1

2

3

a

2

b c d e f g

lg E

', M

Pà

Temperature, °C

1 a)

-60 0 60 120

0,1

0,2

0,3

0,4

à b c d e f g

tan

δ

Temperature, °C

1

2

b)

Figure 26. Temperature dependences of mechanical loss tangent for (1) HDPE and (2) EPDM and for initial TPE and TPE-1–TPE-6 samples (a–g, respectively). samples slightly increases with the increasing number of processing cycles. For example, the glass transition temperature of TPE-4 is by 7 °C higher than that of the initial TPE. This is likely to be related to the reduced segmental mobility of polymer chains in the amorphous phase of TPE. In the TPEs under study, this behavior can be a result of either increased density of the network of physical (crystallites) or chemical (transverse) bonds, or of the improved compatibility between crystalline (HDPE-pc) and mixed amorphous phases (HDPE-pc–EPDM–GTRr). Taking into account the fact that the degree of crystallinity of the TPE-1–TPE-5 samples is lower than that of the initial TPE (Table 14) and the content of gel fraction slightly increases (by 1.0–3.5 %, Fig. 24), glass transition temperature Tg of the amorphous phase in the TPE-1–TPE-5 samples is likely to increase due to the reduced phase separation between the crystalline and amorphous phases. This conclusion is also proved by a slight increase in the intensity of the peaks observed for the TPE-1–TPE-5 samples, as compared with the initial TPE. Therefore, the results of DMTA studies show that, with the increasing number of processing cycles, viscoelastic characteristics of TPEs are changed due to a decrease in the degree of crystallinity of the HDPE matrix and due to an increased crosslinking density. In this case, compatibility between crystalline (HDPE) and amorphous (HDPE–EPDM–GTR) phases in TPEs is improved. Conclusions The effect of compatibilization have been observed for LDPE-pc/BR TPEs prepared in the presence of the following reactive couples: PB-NH2/PE-


co-GMA, PB-NCO/PE-co-AA and PB-NCO/PE-co-VA-co-AA. The most effective reactive couple was PB-NCO/PE-co-AA and the higher effect of compatibilization was reached at 7.5 % of PB-NCO content per BR when the ratio of functional groups for PB- and PE-based modifiers was kept 1/1. In such a case modified TPE obtained has values of TS and EB higher by 31% and 63%, respectively, than for the non-modified LDPE-pc/BR TPE. The results obtained are explained by realization of reaction between PE-co-AA and PB-NCO in polyolefin/rubber interface and, as a result, by improving interfacial adhesion. In modified TPEs shift and convergence of α-relaxation transitions of components toward one another is fixed by DMTA that confirms the interaction of BR and LDPE-pc due to the formation of the essential interface layer. WAXS investigation has shown that presence of BR and reactive modifiers do not prevent completely the crystallization process of LDPE-pc at TPEs formation. Depression of Tm has been found for all TPEs studied. DMTA data have shown that some increase of chain mobility in amorphous phases is observed for modified TPEs in comparison to unmodified TPE or pure LDPE-pc obviously due to increasing of unsoundness of crystalline phase of LDPE-pc. It was found that all TPEs studied are characterized by microphase separation of components and have complicated multiphase structure. The comparative analysis of the results obtained by various techniques allows concluding of transformation of the morphology of the basic blend (continuous LDPE-pc phase and dispersed BR phase) to the morphology with the essential interface layer that provides an improvement of mechanical characteristics of modified TPEs produced. The LDPE-pc (PE-co-AA)/BR (PB-NCO) TPEs studied can be classified as so-called semi-interpenetrating polymer networks (semi-IPNs). Based on this study devoted to produce thermoplastic elastomers, TPEs, using ground tire rubber, GTR, the following conclusions can be drawn:

Bitumen is a suitable reclaiming agent for GTR under the treatment conditions used. During further melt processing it acts as a curing agent for the rubber components of TPEs and works also as an effective compatibilizer for HDPE-pc/EPDM/GTR compositions.

TPEs containing GTR pretreated by bitumen show outstanding mechanical properties, high thermal stability and good reprocessability. In addition, TPE grades containing GTR and recycled HDPE-pc can be produced batch wise and continuously in industrial scale.

The performance of TPEs mainly depends on conditions of the GTR reclaiming by bitumen, type of the rubber used, and melt processing parameters.


Structure-property relationship investigations of the TPEs carried out by SEM, DSC and DMTA methods clearly evidence the improvement in interfacial adhesion between the GTR particles and surrounding thermoplastic matrix when the GTR was partially devulcanized in bitumen.

Analysis of the viscous flow characteristics of TPEs shows that, independently of the number of processing cycles, all samples are characterized by the required flow characteristics at elevated temperatures. Processing of TPEs is accompanied by the competing processes of crosslinking and degradation of macromolecules in a polymer mixture. The results of DSC study and dynamic thermal analysis show that, as the number of processing cycles is increased, phase separation between amorphous and crystalline phases in TPEs decreases. Insignificant intermolecular crosslinking induced by the processing of TPEs appears to have almost no effect on the physicomechanical characteristics of the final material.

References 1. Karger-Kocsis, J. 1999, In: Shonaike GO, and Simon GP, editors. Polymer

Blends and Alloys. New York: Marcel Dekker, Chap. 5. 2. Abdou-Sabet, S. 1998, Polym. Mat. Sci. Eng., 79, 86. 3. George, J., Varughese, K.T., and Thomas, S. 2000, Polymer, 41, 1507-17. 4. Michael, H., Scholz, H., and Menning, G. 1999, Kautch. Gummi Kunstst., 52,

510-3. 5. Yang, Y., Chiba, T., Saito, H., and Inoue, T. 1998, Polymer, 39, 3365-72. 6. Cavalieri, F., Padella, F., and Cataldo, F. 2003, J. Appl. Polym. Sci., 90, 1631-8. 7. Grigoryeva, O., Fainleib, A., Starostenko, O., Danilenko, I., Kozak, N., and

Dudarenko, G. 2004, Rubber Chem. Tech., 77, 131-46. 8. Nevatia, P., Banerjee, T.S., Dutta, B., Jha, A., Naskar, A.K., and Bhowmick,

A.K. 2002, J. Appl. Polym. Sci., 83, 2035-42. 9. McKirahan, J., Liu, P., and Brilhart, M. 1996, ANTEC’96, Society of Plastics

Engineering. 10. Bhattacharya, S.N., and Sbarski, I. 2002, Plastics, Rubber and Composites, 27, 317. 11. Fuhrmann, I., and Karger-Kocsis, J. 2003, J. Appl. Polym. Sci., 89, 1622-30. 12. Naskar, A.K., Bhowmick, A.K., and De, S.K. 2001, Polym. Eng. Sci., 41, 1087-98. 13. Adhikari, B., De, D., and Maiti, S. 2000, Prog. Polymer. Sci., 25, 909-48. 14. Rudheshkumar, S., and Karger-Kocsis, J. 2002, Plastics, Rubber and Composites,

31, 1-4. 15. Rudheshkumar, S., Fuhrmann, I., and Karger-Kocsis, J. 2002, Polym. Degr.

Stab., 76, 137-44. 16. Grigoryeva, O.P., Fainleib, A.M., Tolstov, A.L., and Starostenko, O.N. 2004,

Nonlinear Optics. Quantum Optics, 31, 185-198.


17. Fainleib, A.M., Grigoryeva, O.P., Tolstov, A.L., and Starostenko O.N. 2004, In Composites Technologies for 2020, Cambridge: Woodhead Publishing Ltd, England, 175-180.

18. Corley, B., and Radusch, H.-J. 1998, J. Macromol. Sci.-Phys. B., 37, 265-73. 19. Orr, C.A., Cernohous, J.J., Guegan, P., Hirao, A., Jeon, H.K., and Macosko,

C.W. 2001, Polymer, 42, 8171-8. 20. Li, Y., Zhang, Y., and Zhang, Y.X. 2003, Polym. Test., 22, 859-65. 21. Li, Y., Zhang, Y., and Zhang, Y.X. 2003, Polym. Test., 23, 83-90. 22. Grigoryeva, O., Fainleib, A., Starostenko, O., Tolstov, A., and Brostow, W. 2004,

Polym. Int., 53, 1693-1703. 23. Final report on EU INCO-COPERNICUS project RECRUPOT (contract n.

ICA2-CT-2001-10003), April, 2004. 24. Krimm, S., and Tobolsky, A. 1961, J. Polym. Sci., 7, 57-76. 25. Hermans, P.H., and Weidinger, A. 1961, Makromol. Chem., 44/46, 24-36. 26. Radonjič, G., and Gubeljak, N. 2002, Macromol. Mater. Eng., 287, 122-32. 27. Mark, J.E. 1996, Physical Properties of Polymers. Handbook, Woodbury: ATP

Press, American Institute of Physics, p.723. 28. Grigoryeva, O., Fainleib, A., Tolstov, A., Starostenko, O., Lievana, E., and

Karger-Kocsis, J. 2005, J. Appl. Polym. Sci., 95, 659-671. 29. Fainleib, A.M., Grigoryeva, O.P., Starostenko, O.N., Tolstov, A.L Danylenko,

I.Y., Sergeeva, L.M., and Lebedev, E.V. 2004, Pat. Ukraine. N63593. 30. Fainleib, A.M., Grigoryeva, O.P., Starostenko, O.N., Tolstov, A.L., Danilenko,

I.Y., Sergeeva, L.M., and Lebedev, E.V. 2004, Pat. Russia, N2241720. 31. Fainleib, A., Grigoryeva, O., Starostenko, O., Danilenko, I., and Bardash, L.

2003, Macromol. Symp., 202, 117-126. 32. Greco, R., Musto, P., Riva, F., and Maglio, G. 1989, J. Appl. Polym. Sci., 37, 789-801. 33. McCrum, N.G., Read, B.E., and Williams, G. 1967, Anelastic and Dielectric

Effects in Polymeric Solids, New York: Wiley, p. 355. 34. Grigoryeva, O., Tolstov, A., and Fainleib, A. 2004, Scientific Israel -

Technological Advantages, 6, 37-44. 35. Grigoryeva, O.P., Lievana, E., Tolstov, A.L., Starostenko, O.N., Fainleib, A.M.,

and Karger-Kocsis, J. 2004, In Composites Technologies for 2020, Cambridge: Woodhead Publishing Ltd, England, 181-186.

36. Fainleib, A.M., Tolstov, A.L., Grigoryeva, O.P., Starostenko, O.N., Danylenko, I.Y., Bardash, L.V., and Gafurov, U.G. 2006, Pat.Ukraine, N75849.

37. Lievana, E., and Karger-Kocsis, J. 2004, Prog. Rubber, Plastics and Recycling Technol., 20, 1-10.

38. Mamunya, Ye.P., Bey, I.N., Tolstov, A.L., Fainleib, A.M., and Gladkiy, E.P. 2005, Polymernyy J. (Kiev), 27, 117-122.

39. Grigoryeva, O., Fainleib, A., Tolstov, A., Pissis, P., Spanoudaki, A., Vatalis, A., and Delides, C. 2006, J. Therm. Anal. Calorim., 86, 229-233.

40. George, J., Ramamurthy, K., Varughese, K.T., and Thomas, S. 2000, J. Polym. Sci. Part B: Polym. Phys., 38, 1104-22.

41. Munstedt, H. 1981, Polym. Eng. Sci., 21, 259-70. 42. White, J.L., Czarneck, L., and Tanaka, H. 1980, Rubber Chem. Technol., 53,

823-35.


43. Metzner, A.B. 1985, J. Rheol., 29, 739-45. 44. Mark, J.E. 1996, Physical Properties of Polymers Handbook. Woodbury: American

Institute of Physics Press. 45. Akovali, G., and Torun, T.T. 1997, Polym. Int., 42, 307-14. 46. Schatzki, T. 1962, J. Polym. Sci., 57, 496. 47. Grassie, N. 1956, Chemistry of High Polymer Degradation Processes. London:

Butterworths. 48. Grigoryeva, O.P., Fainleb, A.M., Shumskii, V.F., Vilenskii, V.A., Kozak, N.V.,

and Babkina, N.V. 2009, Vysokomol. Soed., Ser. A, 51, 275–285. 49. Jansson, A., Moller, K., and Gevert, T. 2003, Polym. Degrad. Stab., 82, 37-46. 50. Runt, J.P. 2000, In: Paul DR, and Bucknall CB, editors. Polymer Blends. New York:

Wiley, Vol. 1. 51. Vasile, C., and Kulshreshtha, A.K. 2003, In: Handbook of Polymer Blends and

Composites. Shawbury: Rapra Technology, Vol. 3B.

1. utilization of tire rubber and recycled polyolefins into

Documents