EVA

http://www.e-polymers.org

e-Polymers 2012, no. 005

ISSN 1618-7229

Dual phase continuity and phase inversion phenomena in natural rubber/ethylene vinyl acetate (EVA) copolymer blends Wisut Kaewsakul, Azizon Kaesaman, Charoen Nakason* Center of Excellence in Natural Rubber Technology, Dept Rubber Technology & Polymer Science, Faculty of Science & Technology, Prince of Songkla University, Pattani, Thailand; fax: 66 73 335130; e-mail: [email protected] (Received: 23 May, 2010; published: 22 January, 2012) Abstract: Blending of polymers provide an attractive route to obtain new materials with unique combination of properties. Among the possible blend morphologies are dispersion of droplet, fiber, lamella in continuous phase and cocontinuous morphology. The unique feature of co-continuous is the mutual interpenetration of the phases which exhibits a number of advantageous properties and hence wider range of applications. Therefore, it is important to perform accurate determination of the co-continuous phase morphology and phase inversion phenomena. The rubber co-continuity index (solvent extraction), SEM micrographs and dynamic properties indicated the inversion occurred in the blends with rubber content 40 to 50 wt%. Furthermore, three empirical models (i.e., Ultracki, Steiman et al and Gergen et al models) could predict the phase continuity and phase inversion phenomena but other two (i.e., Paul-Barlow and MetelkinBlekht models) failed. Therefore, based on rubber co-continuity index, SEM micrographs, dynamic mechanical properties and prediction, the inversion phenomena occurred in the blends with rubber contents in a range of 40-50 wt%. Full co-continuity was found in the blends with the rubber contents of 46 to 60 wt%.

Introduction Blending of polymers provide an attractive route to obtain new materials with unique combination of properties. The blending of different polymer pairs usually gives rise to new materials with properties that cannot be achieved from the individual components. Thus, a number of research works have been focused on enhancing properties of the blend materials for a specific application and for better combination of properties. Thermoplastic elastomers (TPEs) based on rubber and thermoplastic blends are such materials which combine physical properties of conventional thermoset rubber and excellent processing characteristics of thermoplastic. Among the possible blend morphologies are dispersion of droplet, fiber, lamella in continuous phase and co-continuous morphology [1]. The unique feature of co-continuous is the mutual interpenetration of the phases. Polymer blend with co-continuous phase morphologies exhibits a number of advantageous properties and hence wider range of applications [2]. In elastomer/thermoplastic blend without curative, only the blends with co-continuous phase morphologies exhibit the thermoplastic elastomer characteristics [3]. Therefore, reactive blending and the blend with compatibilizer are normally used to prepare the TPEs with partly compatibilized blends. This is to enhance interfacial adhesion between the interfaces and hence other related physical properties [4-6]. Therefore, it is important to perform accurate determination of the co-continuous phase morphology to verify the occurrence of thermoplastic 1 Unauthenticated Download Date | 9/24/15 11:29 PM

elastomer materials based on rubber and plastic blends. A variety of methods have been used to determine co-continuity including microscopy, solvent extraction, rheological measurements and electrical conductivity [7-11]. Thermoplastic elastomers based on natural rubber and thermoplastic blends are classified as thermoplastic natural rubber (TPNR). Various types of thermoplastics have been used to prepare TPNRs. These include polypropylene [12-14], polyethylene [15,16], polystyrene [17], polyamide-6 [18], and poly(methyl methacrylate) [19]. Ethylene-vinyl acetate copolymer (EVA) has also been used to prepare the TPNRs [20-23] due to its excellent ozone damage resistance, aging ability, weather resistance, and mechanical properties [24]. Moreover, EVA is a halogen-free thermoplastic which may be used as a replacement in many applications that currently are dominated by PVC. Apart from raw NR, modified forms of NR have also been extensively used to prepare TPNRs. These include epoxidized natural rubber (ENR) [25-27], maleated natural rubber (MNR) [28,29], and natural rubber-g-poly(methyl methacrylate) [30-32]. In this work, an attempt was made to detect the phase continuity and phase inversion phenomena of natural rubber/EVA blends. Three different forms of NR were exploited: unmodified NR, air dried sheet (ADS), maleated natural rubber (MNR) and epoxidized natural rubber (ENR). Phenolic modified EVA (PhHRJ-EVA) was used as a blend compatibilizer in the ADS/EVA blend to enhance the blend compatibility. However, the reactive blends of ENR/EVA and MNR/EVA were prepared without blend compatibilizer. Results and discussion Solvent extraction Figure 1 shows the co-continuity index of rubber phase as a function of rubber content in the blends. It is seen that the compositions at which the different blends exhibit full co-continuous phase morphologies cover a wide range of blend proportions.

Fig. 1. Rubber phase co-continuity index with various rubber contents of rubber/EVA blends and three different types of natural rubber. 2 Unauthenticated Download Date | 9/24/15 11:29 PM

The rubber phase co-continuity threshold with values of 47.48, 68.81, and 65.31 for ADS/EVA, MNR/EVA and ENR/EVA with a blend ratio of 20/80 were observed, respectively. At this composition, the blend is composed of droplets of rubber phase dispersed in the EVA matrix. In Figure 1, it is also seen that the blends exhibit full cocontinuous phase morphology in the composition ranges of rubber from 50 to 60 wt%. Increasing higher content of rubber than 60 wt%, the phase inversion may occur, as a decreasing trend of the rubber co-continuity index was observed. The phase morphology of the blends with rubber content = 40 wt% approached to the cocontinuity. This is due to the rubber co-continuity index was greater than 70%. However, to clarify the precise phase morphology of this blend composition, other means of characterization are needed. The phase morphology developed in three types of natural rubber/EVA blends was also followed by SEM micrographs and dynamic properties. Scanning electron microscope To confirm the phase continuity of the blends, scanning electron microscopy (SEM) was performed on cryofractured gold-coated surfaces using a scanning electron microscope. Molded sheets of the blends were cryogenically cracked in liquid nitrogen and then rubber phase was extracted by hexane for ADS/EVA blend and chlorobenzene for MNR/EVA as well as ENR/EVA blends.

Fig. 2. SEM micrographs of ADS/EVA blends with various blend proportions.

3 Unauthenticated Download Date | 9/24/15 11:29 PM

Figures 2-4 show SEM micrographs of ADS/EVA, MNR/EVA and ENR/EVA blends with various blend ratios. It can be seen that the rubber phase was extracted and shown as cavitations in the remaining EVA phases. This confirms two-phase morphology with the rubber droplets dispersed in the EVA matrix at low compositions of rubber at a blend ratio of rubber/EVA = 20/80.

5 µm

5 µm

(A) MNR/EVA = 20/80

(B) MNR/EVA = 40/60

5 µm

(C) MNR/EVA = 50/50

(D) MNR/EVA = 60/40

5 µm

(E) MNR/EVA = 80/20 Fig. 3. SEM micrographs of MNR/EVA blends with various blend proportions. Increasing rubber contents to 40 wt% caused the phase inversion phenomena to cocontinuous phase morphology in particular at the blend ratios of rubber/EVA = 50/50 and 60/40 wt%. It can be clearly seen that the patterns of both phases are similar and interpenetrated into each others. The SEM result of the blend with rubber/EVA = 40/60 showed the transition from dispersion to the co-continuous phase morphology. This confirms the results of the phase co-continuity index of rubber phase in Figure 1. Furthermore, increasing content of rubber higher than 60 wt% (i.e., rubber/EVA = 4 Unauthenticated Download Date | 9/24/15 11:29 PM

80/20), the second phase inversion phenomena occurred where the EVA dispersed in rubber matrix.

Fig. 4. SEM micrographs of ENR/EVA blends with various blend proportions. Dynamic properties The phase continuity and phase inversion phenomena in natural rubber/EVA blends were also proved by the dynamic properties in terms of relationship between tan δ at a frequency of 1 Hz and weight fraction of rubber in the blends, as shown in Figure 5. It is seen that the pure EVA (i.e., weight fraction of rubber = 0 wt%) showed the highest tanδ value indicating higher dissipated energy or loss modulus compared with the elastic modulus. Furthermore, the tan δ values decreased when incorporation of rubber and exhibited decreasing trend with increasing rubber content in the blends. This is attributed to lower dissipated energy nature of the rubber. Further increased rubber contents higher than 40% caused abruptly decreasing trend of the tan δ until reached slight decreasing trend in the blend with rubber content of approximately 60 wt%. This is attributed to the phase inversion from droplets of rubber particles dispersed in the EVA matrix to co-continuous phase morphology 5 Unauthenticated Download Date | 9/24/15 11:29 PM

where both phases behave as continuous phases. Lower dissipated energy and hence higher elasticity of the blends was observed for the co-continuous phase materials. According to this result, it is seen that the first phase inversion may occur at the abruptly changed of the tan δ where the rubber contents are in the range 40 to 50 wt%. This confirms the phase inversion phenomena and co-continuity based on extraction experiment (Figure 1) and SEM results (Figures 1-3). It is also noted that in Figure 5, the ENR/EVA exhibited the lowest tan δ and the ADS/MNR showed the highest value. Furthermore, the MNR/EVA exhibited the intermediate values. This is attributed to the reactive blending of ENR and EVA caused the strongest interaction between the polar functional groups of ENR and EVA. However, the reactive blending of MNR and EVA caused weaker interaction between the phases. On the other hand, the ADS/EVA blends with the blend compatibilizer caused the lowest interaction between NR and EVA. Details of reactive blending for these types of blends and the use of blend compatibilizer were described elsewhere [34].

Fig. 5. Tan δ at a frequency of 1 Hz of rubber/EVA blends with different types of natural rubber and blend proportions. Phase inversion composition and co-continuity detection There have been several models used to estimate the phase inversion composition and co-continuity based on material properties and processing conditions. The prediction has been normally used to verify the phase inversion composition and co-continuity detection compared with the experimental results in terms of extraction experiment, microscopy images, mechanical and dynamic properties of the blends. According to the empirical relationship for prediction of the phase inversion suggested by Avgeropoulos [35] and generalized by Paul and Barlow [36], the phase inversion occurs when the volume ratio is equal to the viscosity ratio between the blend components. In the NR/EVA blend system, this relationship is described as follows:


φ NR =

1

(1)

η (1 + EVA ) η NR

where η and φ denote the viscosity and volume fraction of the blend components, respectively. Utracki [37] developed a model based on emulsion theory which included intrinsic viscosity in the equation. In the NR/EVA blend system, the Ultraki’s model leads to an expression:

φ NR

⎛ ⎞ ⎛η ⎞ ⎜1 − log⎜ EVA ⎟ /[η ]⎟ ⎜η ⎟ ⎜ ⎟ ⎝ NR ⎠ ⎝ ⎠ = 2

(2)

where the intrinsic viscosity was assumed to be 1.9 for spherical domains. Metelkin and Blekht [38] proposed another approach using a concept of capillary instabilities of individual layers. The model was applied to the NR/EVA blend system, as follows:

φNR

2 ⎡ ⎡ ⎛ ⎛ η EVA ⎞ ⎞ ⎤ ⎤ ⎛ η EVA ⎞ η EVA ⎢1 + 2.25 log⎜⎜ ⎟ + 1.81⎜ log⎜⎜ ⎟⎟ ⎟ ⎥ ⎥ = ⎢1 + ⎜ ⎟ η NR ⎟⎠ η ⎢ η NR ⎢ NR ⎝ ⎝ ⎠ ⎠ ⎥⎦ ⎥⎦ ⎝ ⎣ ⎣

−1

(3)

Steinmann et al. [39] proposed an approach based on the assumption that the shape relaxation times of blend components meet a maximum at the phase inversion composition. They found a strong correlation between the viscosity and elasticity ratios. A corresponding equation was given based on the viscosity ratio at a certain constant elasticity. Therefore, according to this model, the phase inversion for the NR/EVA blend system can be written as ⎛ η EVA ⎞ ⎟⎟ + 0.48 ⎝ η NR ⎠

φNR = −0.12 log⎜⎜

(4)

Gergen et al [40] also proposed a relationship between ratio of viscosities and the proportions of the two components. It can be expressed for the blend of NR and EVA as follows. log(ηEVA/ηNR) =

2-4φNR

(5)

The prediction of phase inversion composition and co-continuity according to the above five models was performed based on the complex viscosities of pure EVA and different forms of NR (i.e., ADS, MNR and ENR) in a range of frequency of 0.4 to 190 rad/s and the viscosity ratios in a range of 0.1-10, as the complex viscosities results shown in Figure 6. The shear rate or frequency generated in the internal mixer at a rotor speed of 60 rpm at 140 °C has been approximated to 100 rad/s. Therefore, the viscosity ratios of φEVA/φADS, φEVA/φMNR and φEVA/φENR = 0.40, 0.73 and 0.52 can be estimated at a frequency of 100 rad/s, respectively. This can be assumed with a good approximation to be the average shear rates in the internal mixer at a given rotor speed of 60 rpm.


Fig. 6. Complex viscosity of different forms of natural rubber and EVA. Figures 7-9 show the prediction of phase inversion composition and co-continuity according to the above five models. Based on the predicted results in Figures 7-9, at the viscosity ratios of 0.40, 0.73 and 0.52 and using eqs. (2)-(6), the phase inversion point or the phase boundary at which the morphology first changes from NR dispersed in the EVA matrix to the co-continuous morphology is concluded in Table 1.

Fig. 7. The predicted phase inversion volume fraction as a function of the viscosity ratios of ADS/EVA blends.


Fig. 8. The predicted phase inversion volume fraction as a function of the viscosity ratios of MNR/EVA blends.

Fig. 9. The predicted phase inversion volume fraction as a function of the viscosity ratios of ENR/EVA blends. It is seen that the three models (Ultracki, Steiman et al and Gergen et al models) could determine the phase boundary where the dispersion of rubber changed to cocontinuous phase morphology at weight fractions of rubber in the ranges of 46-55, 45-50 and 46-54 wt%, respectively. Therefore, based on these models, the first 9 Unauthenticated Download Date | 9/24/15 11:29 PM

phase inversion was estimated at the rubber content = 46 wt%. The results agree with the experimental results based on extraction experiment (Figure 1) and morpological properties (Figures 2-4) and dynamic properties (Figure 5). However, Paul-Barlow and Metelkin-Blekht models failed to predict these types of blends due to the phase inversion boundary with very high content of rubber was observed. Therefore, it is concluded that the first phase inversion occurs at the blend ratio of rubber/EVA approximately = 46/54 wt %. The prediction, using these models, is not fully understood yet, and there is a real need for other complimentary studies. The inversion position after the co-continuity with the experimental result at a blend ratio higher than 60/40 wt% cannot be determined by these models. Tab. 1. Prediction of phase inversion compositions for NR/EVA blends at viscosity ratios calculated based on complex viscosity at a frequency of 100 rad/s. Models

ADS/EVA Weight fraction of rubber (%) 0.72 65 0.61 55 46 0.53 0.60 54 0.87 78

φADS

Paul-Barlow [36] Utracki [37] Steinman et al. [39] Gergen et al. [40] Metelkin-Blekht [38]

MNR/EVA Weight fraction of rubber (%) 0.58 54 0.54 50 0.50 45 0.53 50 0.65 62

φMNR

ENR/EVA Weight fraction of rubber (%) 0.66 62 0.57 54 0.51 46 0.57 54 0.79 74

φENR

Conclusions Based on rubber co-continuity index from solvent extraction, SEM micrographs, dynamic mechanical properties and prediction, it was found that the first inversion occurred in the blends with rubber content in a range of 40-50 wt%. Furthermore, the fully co-continuity was observed in the blends with the rubber contents of 50 to 60 wt%. Three models (Ultracki, Steiman et al and Gergen et al models) could predict the phase boundary where the dispersion of rubber changed to co-continuous phase morphology at weight fractions of rubber in the ranges of 46-55, 45-50 and 46-54 wt%, respectively. However, Paul-Barlow and Metelkin-Blekht models failed to predict the blends due to the phase inversion boundary with very high content of rubber was observed. Experimental part Materials Natural rubber, air dried sheet (ADS) was manufactured by the Khuan Pun Tae Farmer Co-operation, (Phattalung, Thailand). Maleated natural rubber (MNR) was prepared in-house by melt blending of ADS and 10 phr of maleic anhydride at 140 oC and a rotor speed of 60 rpm. The details of preparation and characterization procedures of MNR were described elsewhere [29]. Epoxidized natural rubber (ENR) with a level of epoxide group of 35 mole % was also prepared in-house using high ammonia (HA) concentrated NR latex via performic epoxidation. The preparation and characterization procedures of the ENR were described in our previous work [33]. Ethylene-vinyl acetate copolymer (EVA, grade N8308) used as a blend component having 18 wt % of vinyl acetate content with MFI of 2.3 g/10 min (2.16 kg/190 oC) and 10 Unauthenticated Download Date | 9/24/15 11:29 PM

density of 0.941 g/cm3. It was manufactured by Thai Polyethylene, Co., Ltd, (Rayong, Thailand). Phenolic resin with active hydroxymethyl (methylol) groups (grade HRJ10518) used to prepare a blend compatibilizer [phenolic modified EVA (PhHRJ-EVA)] was manufactured by Schenectady International Inc., (New Port, USA). The PhHRJEVA was prepared in-house by melt mixing EVA, phenolic resin and stannous chloride at 140 oC and rotor speed of 60 rpm, as the procedure described elsewhere [3]. Stannous chloride used as a catalyst in preparation of PhHRJ-EVA, was manufactured by Fluka Chemie (Buch, Switzerland). Blend preparation Three types of polymer blends based on EVA with three different forms of natural rubber (i.e., ADS/EVA, MNR/EVA and ENR/EVA blends) were prepared with various blend ratios of rubber/EVA = 20/80, 40/60, 50/50, 60/40 and 80/20. Blending was carried out by melt mixing process using an internal mixer with intermeshed rotors (Charoen-Tad, Co., Ltd., Bangkok, Thailand) with a mixing capability of 500 cm3. The thermoplastic component, EVA, was first dried in hot air oven at 40 oC for at least 10 h to eliminate moisture content. The material was then introduced into the mixing chamber. Mixing was performed for 2 min at a rotor speed of 60 rpm at 140 oC. The rubber component was then incorporated into the mixing chamber and mixing continued for another 4 min. In the ADS/EVA blend, the PhHRJ-EVA compatibilizer at a loading level of 5 wt % of EVA was incorporated into the mixing chamber and mixed for 1 min before blending with rubber. The blending products were cooled down to room temperature and then sheeted out by two-roll mill. The compression molded sheets with a thickness of approximately 2 mm were then prepared at 120 °C, 5 min and then cooled down by water circulating system. Dynamic properties Dynamic properties of the natural rubber/EVA blends were characterized using a rotorless oscillating shear rheometer (rheo TECH MDPT, Cuyahoya Falls, USA) using a frequency mode with a range of oscillation frequency of 0.4 to 190 rad/s with a constant strain of 7 % and temperature of 140°C. The storage (G′) and loss shear (G″) moduli, loss factor, tan δ = G″/G′ as well as the complex viscosity (i.e., η* = G*/ ω = η″ + iη′) of various types of the blends were characterized. Scanning electron microscopy To detect phase continuity of the blends, scanning electron microscopy (SEM) was performed on cryofractured gold-coated surfaces using a scanning electron microscope, model JSM-5200 (Jeol Co. Ltd., Japan). Moulded sheets of the blends were first cryogenically cracked in liquid nitrogen to avoid any possibility of phase deformation during the cracking process. The fractured surfaces were etched by solvents (i.e., hexane for ADS/EVA blend, chlorobenzene for MNR/EVA and ENR/EVA blends) at room temperature for 48 h to remove the rubber phase. The samples were later dried in a vacuum oven at 40 °C for 3 h to eliminate contamination of the solvent and then gold-coated before examining by SEM. Solvent extraction Solvent extraction experiments were used to determine the phase composition where the blends exhibit dispersed phase morphology or where the phases are co11 Unauthenticated Download Date | 9/24/15 11:29 PM

continuous. Samples with regular thickness of approximately 1 mm and known weight (20-40g) for each blend were stirred in solvents (i.e., hexane for ADS/EVA blend, chlorobenzene for MNR/EVA and ENR/EVA blends) at 30 °C for 7 days to selectively dissolve the rubber phase. The samples were then removed from solution, and each solution was filtered using a 0.7 µm Whatman FilterCup GF/F. The samples and filters were dried in a vacuum oven at 40 °C for 48 h. The mass of the sample after extraction was determined. The extraction process was repeated until a constant sample mass was observed. The continuity index of the rubber phase (CIrubber) was calculated as the percentage of the rubber phase that was extracted using the following relation: [8]

CI rubber (%) =

mini − mext mini

x

(6)

100

where mini is the weight of the rubber phase initially present in the blend and mext is the weight of the rubber phase in the blend after solvent extraction. Acknowledgements The authors thank Prince of Songkla University for financial support and research facilities and other support, contract no. SAT50109900099S. This work was also supported by the Higher Education Research Promotion and National Research University Project of Thailand, Office of the Higher Education Commission.900099S. References [1] Macosko, C.W. Macromol. Symp, 2000, 149, 171. [2] Liu, Z.H.; Marechal, P.; Jerome, R. Polymer, 1978, 39, 1779. [3] Nakason, C.; Nuansomsri, K.; Kaesaman, A.; Kiatkamjornwong, S. Polym Test, 2006, 25, 782. [4] Mohamad, Z.; Ismail, H.; Thevy, R.C. J Appl Polym Sc, 2006, 99, 1504. [5] Nakason, C.; Saiwaree, S.; Tatun, S.; Kaesaman, A. Polym Test, 2006, 25, 656. [6] Pechurai, W.; Nakason, C.; Sahakaro K. Polym Test, 2008, 27, 621. [7] Galloway, J.A.; Koester, K.J.; Paasch, B.J.; Macosko, C.W. Polymer, 2004, 45, 423. [8] Omonov, T.S.; Harrats, C.; Moldenaers, P.; Groeninckx, G. Polymer, 2007, 48, 5917. [9] Pechurai, W.; Nakason, C.; Sahakaro, K. Polym Test, 2008, 27, 621. [10] Galloway, J.A.; Montminy, M.D.; Macosko, C.W. Polymer, 2002, 43, 4715. [11] Steinmann, S.; Gronski, W.; Friedrich, C. Polymer, 2001, 42, 6619. [12] Riahi, F.; Benachour, D.; Douibi, A. Int J Polym Mat, 2004, 53, 143. [13] Varghese, S.; Alex, R.; Kuriakose, B. J Appl Polym Sci, 2004, 92, 2063. [14] Jeong Seok, O.H.; Isayev, A.I.; Rogunova, M.A. Polymer, 2003, 44, 337. [15] Dahlan, H.M.; Zaman, M.D.K.; Ibrahim, A. J Appl Polym Sci, 2000, 78, 1776. [16] Abdullah, I.; Ahmad, S.; Sulaiman, C.S. J Appl Polym Sci, 1995, 58, 1125. [17] Asaletha, R., Kumaran, M.G.; Thomas, S. Eur Polym J, 1999, 35, 253. [18] Carone, Jr E.; Kopcak, U.; Gonc-alves, M.C.; Nunes, S.P. Polymer, 2000, 41, 5929. [19] Mina, M.F.; Ania, F.; Calleja, F.J.B.; Asano, T. J Appl Polym Sci, 2004, 91, 205. [20] Koshy, A.T.; Kuriakose, B.; Thomas, S.; Varghese, S. Polymer, 1993, 34, 3428. [21] Mohamad, Z.; Ismail, H.; Thevy, R.C. J Appl Polym Sci, 2005, 99, 1504. 12 Unauthenticated Download Date | 9/24/15 11:29 PM

[22] Bandyopadhyay, G.; Bhagawan, S.; Ninan, K.; Thomas, S. J Appl Polym Sci, 1999, 72, 165. [23] Sujith, A.; Unnikrishnan, G. J. Polym. Res., 2006, 13, 171. [24] Jansen, P.; Gomes, A.S.; Soares, B.G. J Appl Polym Sci, 1996, 61, 591. [25] Mohanty, S.; Nando, G.B.; Vijayan, K.; Neelakanthan, N.R. Polymer, 1996, 37, 387. [26] Nakason, C.; Panklieng, Y.; Kaesaman, A. J Appl Polym Sci, 2004, 92, 3561. [27] Ramesh, P.; De, S.K. J Appl Polym Sci, 1993, 50, 1369. [28] Nakason, C. ; Saiwari, S. ; Tatun, S.; Kasaman, A. Polym Test, 2006, 25, 656. [29] Pichaiyut, S.; Nakason, C.; Kaesaman, A.; Kiatkamjornwong, S. Polym Test, 2008, 27, 566. [30] Oommen, Z.; Thomas, S.; Premalatha, C.K.; Kuriakose, B. Polymer, 1997, 38, 5611. [31] Thiraphattaraphun, L.; Kiatkamjornwong, S.; Prasassarakich, P.; Damronglerd, S. J Appl Polym Sci., 2001, 81, 428. [32] Nakason, C.; Pechurai, W.; Sahakaro, K.; Kaesaman, A. Poly Adv Technol, 2005, 16, 592. [33] Nakason, C.; Wannavilai, P.; Kasaman, A. Polym Test, 2006, 25, 34. [34] Kaewsakul, W. Thermoplastic elastomer based on different forms of natural rubber and EVA blends, MSc Thesis, Prince of Songkla University, Thailand, 2009. [35] Avgeropoulos, G.N.; Weissert, F.C.; Biddison, P.H.; BÖhm, G.G.A. Rubb. Chem. Technol., 1976, 49, 94. [36] Paul, D.R.; Barlow, J.W. Macromol. Sci, Rev. Macromol. Chem., 1980, C18(1), 109. [37] Utracki, L.A. Polym. Mater. Sci. Eng., 1991, 65, 50. [38] Metelkin, V.I.; Blekht, V.I. Colloid J. USSR., 1984, 46, 425. [39] Steinmann, S.; Gronski, W.; Friedrich, C. Polymer, 2001, 42, 6619. [40] Gergen, W.P.; Lutz, R.G.; Davisson, S. Thermoplastic Elastomer, In Holden, G.; Legge, N.R.; Quirk, R.P,; Schroeder, H.E., editors, Hanser Publisher, Munich, 2nd ed. (Chapter 11) 1996, 297.
