Received  Jun. 28, 2001

1 INTRODUCTION
In today's polymer industry, the importance of polymer blends is unquestionable. Most polymer blends are immiscible and have a distinct dispersed phase^[1]. When the volume fraction of the dispersed phase increases, at certain concentration a structural transition in a melt flow structure takes place^[2]. The studies on emulsions of oil and water are an example of such a phase inversion. In the oil-water system there exists a narrow interval volume fraction where the co-continuous phase structure can be observed. As for the polymer blends, the phase inversion interval is known to be quite wide. The only operationally applicable definition of the phase inversion point has been given in many literatures^[2,3]. Veenstra et al.^[4] have shown that co-continuous morphologies are not formed at a point of phase inversion, but rather over a range of volume fractions which strongly depends on the processing conditions and the rheological properties of the components. Lyngaae-Jorgensen et al.^[2] investigated the critical volume fraction for interpenetrating phase structure formation during flow. Ratnagiri and Scott^[3] studied the effect of rheology of the minor component on its tendency to form a continuous phase at short mixing times, showing that phase inversion can occur even with a low-melting major component during compounding. Since the macroscopic mechanical properties of the blends are strongly dependent on the component properties, compositions, and microstructure such as dispersion of minor component and interfacial phase, it is anticipated that the blends have a clearly change in properties at the phase inversion region^[5]. In view of the morphology-property relation, the present work is concerned with the determination of the phase inversion in terms of the immiscible blends polypropylene/polystyrene(PP/PS) by means of the SEM observation, mechanics of materials and rheology.

2 EXPERIMENTAL
PS with molecular weight =5.8×10⁵ and =2.7×10⁵, density 1.05g/cm³, glass transition temperature 111°C (Dynamic Mechanical Analysis) and PP with glass transition temperature 17°C from Beijing Yanshan Petrochemical Company (Beijing, China) were used.
    PS and PP were dried at 80°C for 4 hours and blended in a mixed apparatus (Model XXS-30, China) at 30rpm and 220°C for mixing time 5min. In this processing condition, the shear rate was evaluated nearly 48s^-1. Plates of the blends, PS and PP were compressed in a common heated press at about 20MPa, T=180°C for 1min and then cooled under pressure 20MPa with a resident time of 5min at room temperature. The plates were punched to several dumbbell shape specimens for mechanical tests. Pieces of the plate were broken in liquid nitrogen and the fracture surfaces were etched with acetone with a resident time of 1 min. They were then coated with Au-Pd for observation in a SEM(HITACHI X-650, Japan). Tensile tests were conducted with an Instron Tester (model 1186) at a cross-head speed of 5mm/min and ambient temperature of 25°C. An average of approximately six samples was used for each property.

3 METHOD OF MECHANICS
Many factors such as constitutive material, interfacial properties, and homogeneity of dispersed phase have significantly effect on the strength. The present work is concerned with the effect of component strength and composition on tensile strength of plastics/plastics blends.
    Without loss in generality, the stress-strain curves for two constitutive materials are shown schematically in Fig. 1. It is anticipated that there exists a certain minimum volume fraction f_minof dispersed phase corresponding to the phase inversion point. When the blend has less volume fraction than f_min, the ultimate strength will be controlled by matrix deformation, i.e.,
s^*_U =f_ms^*_m       (1)
where superscript (*) denotes ultimate quantity, subscript m, d denote the matrix and dispersed phase, respectively.
    With the increase of composition of dispersed phase, the morphology will be changed. Once the volume fraction of dispersed phase exceeds the minimum value, we suggest that the blend strength will be transferred to be controlled by the dispersed phase deformation. The corresponding ultimate strength is then expressed by:

s^*_U = f_ds^*_d+f_ms_m     (2)

where s_m is matrix stress at a matrix strain equal to the ultimate strain in the dispersed phase. Combining Equ.(1) with Equ.(2) and applying  f_m+f_d=1 yields the minimum fraction or namely the phase inversion point

f_min=(s^*_m-s_m)/(s^*_d+s^*_m-s_m)    (3)

4 RESULTS AND DISCUSSIONS
Variations of typical morphologies with the ratio of components of the blends of PP/PS are shown in Fig. 2. At low volume fraction of dispersed phase, the morphology has the appearance of particle/matrix. As expected, the co-continuous morphology can be found at the PP/PS(vol%)=60/40, 50/50 and 40/60.
    It is assumed that polypropylene is elastic-perfectly plastic or elastic-plastic. Then the polypropylene would fail at the yield point and its yield strength is considered as the tensile strength. As for the PS, the fracture strength is taken as its tensile strength since the PS fails in a brittle fracture mode at room temperature. The experimental strength and modulus as well as the viscosity at shear rate 48s^-1 and temperature 180° C of the PS and PP were listed in table 1. According to the Hooke's law, the failure strain can be calculated easily, e^*_d=1.0% for the PS and e^*_m=2.7% for the PP. When the matrix strain gets to the ultimate strain e^*_d, the matrix stress s_m will be 9.4MPa. Substitution of s^*_d, s^*_m and s_m into the Equ.(1) and Equ.(2) yields the predicted strengths. Fig. 3 showed the experimental tensile strengths of blends with mixer resident time of 5min as well as the predicted formulas. The PP/PS (60/40) blends had the lowest experimental mechanical strength. As the volume percentage of PS was more than 40%, the mechanical strength was controlled by deformation of PS. The minimum fraction was calculated to be about 35%; this is close to the phase inversion region determined by observing the blend morphologies. It is confirmed that the blend will have stationary mechanical properties in the phase inversion region^[6]. Because the poor interface was not taken into account and an assumption was made to the failure of entire dispersed phase, over estimates were found in Fig. 3. By improving interfacial reaction, one may achieve an expected theoretical strength.
    Moreover, the point of phase inversion was calculated considering the rheological properties of the components, PP and PS as follows^[3],

f_PS /f_PP=  h_PS /h_PP           (4)
with f_PP+f_PS=1

where the subscript denotes the component, h represents the viscosities and f the volume fraction. It should be noted that the long-mixing-time are essential. In this work, 5min of mixing may be long enough, we suggested, to form a steady-state morphology. The point of phase inversion was then computed by the Equ.(4) here about f_PS =0.37, which agrees to the SEM observation and the mechanical prediction. Of these methods the SEM can provide a range of visual phase inversion whereas the others not.

 
        (a) 80/20                                                                  (b) 70/30
 
            (c) 60/40                                                           (d) 50/50

   
               (e) 40/60                                                            (f) 30/70

 
       (g) 20/80

Fig.3 Relation between tensile strength and composition, f_PS =volume percentage of PS

[ Back ] [ Home ] [ Up ] [ Next ] Mirror Site in USA  Europe  China