Fractionated crystallization blends

Ghijsels et al. [1982] investigated the multiple crystallization behavior of blends in which the crystallizable PP phase was finely dispersed into a SBS-rubber (TR). In the case where the latter was finely dispersed, the authors found the PP phase to crystallize at much higher undercooling. A serious drop in the degree of crystalhnity, X, was also reported. The melting behavior of the fractionated crystallized blend did not seem to be markedly affected, e.s., AH and T remained constant, independent of the amount TR added. 

The preparation of immiscible polymer blends is another way to disperse a bulk polymer into fine droplets. It has been reported for several polymers that when they are dispersed in immiscible matrices into droplets with average sizes of around 1 pm, they usually exhibit multiple crystallization exotherms in a differential scanning calorimetry (DSC) cooling scan from the melt (at a specific rate, e.g., 10 Cmin ). Frensch et al. [67] coined the term fractionated crystallization to indicate the difference exhibited by the bulk polymer, which crystallizes into a single exotherm, in comparison with one dispersed in a large number of droplets, whose crystallization is fractionated temperature-wise during cooling from the melt. 

In order to illustrate the fractionated crystallization behavior we will present here previous results on immiscible atactic PS and isotactic polypropylene blends (iPP) [68]. The cooling behavior of PS, iPP and an 80/20 PS/iPP blend is presented in Fig. 1, as well as that of an unmixed blend , labeled... 

The melt mixed 80/20 PS/iPP blend displays a set of exotherms, where the amount of the iPP component that was heterogeneously nucleated is substantially reduced as indicated by the decrease of the crystallization enthalpy in the temperature region where the iPP crystallizes in bulk, i.e., at 109-111 °C (exotherm labeled A). This effect is due to the confinement of iPP into a large number of droplets. If the number of droplets of iPP as a dispersed phase is greater than the number of heterogeneities present in the system, fractionated crystallization occurs. The number of droplets for this composition is known (by scanning electron microscopy observations) to be of the order of 1011 particles cm-3 and polarized optical microscopy (POM) experiments have shown that this iPP contains approximately 9 x 106 heterogeneities cm-3. In fact, it can be seen in Fig. 1 that the fractionated crystallization of the iPP compon-... 

Figure 1 shows the DSC cooling scan of iPP in the bulk after self-nucleation at a self-seeding temperature Ts of 162 °C (in domain II). The self-nucleation process provides a dramatic increase in the number of nuclei, such that bulk iPP now crystallizes at 136.2 °C after the self-nucleation process this means with an increase of 28 °C in its peak crystallization temperature. In order to produce an equivalent self-nucleation of the iPP component in the 80/20 PS/iPP blend a Ts of 161 °C had to be employed. After the treatment at Ts, the cooling from Ts shows clearly in Fig. 1 that almost every iPP droplet can now crystallize at much higher temperatures, i.e., at 134.5 °C. Even though the fractionated crystallization has disappeared after self-nucleation, it should also be noted that the crystallization temperature in the blend case is nearly 2 °C lower than when the iPP is in the bulk this indicates that when the polymer is in droplets the process of self-nucleation is slightly more difficult than when it is in the bulk. In the case of block copolymers when the crystallization is confined in nanoscopic spheres or cylinders it will be shown that self-nucleation is so difficult that domain II disappears. 

In the case of polymer blends, the fractionated crystallization phenomenon that has been widely reported for many polymer systems can not be attributed to simple size effects. For instance, in Fig. 1, one could argue that the different exotherms originated in the crystallization of different droplet populations that have diverse average diameters. This cannot be the case, since the droplet distribution is monomodal and a smooth variation in heat... 

Figure 5 shows the Fisher-Tumbull plot for the crystallization of blends of PS in sesame oil (26, 42, 60, and 80%) (9). This model system is a complex crystallization system. PS is a mixture of TAG obtained through fractional crystallization from refined, bleached, and deodorized palm oil (14). Tripalmitin is the TAG with the highest melting temperature in PS (15), and here its concentration was 16.46% w/w ( 0.17%) (9). Our previous research showed that tripalmitin mostly determines the crystallization kinetics of PS and its blends with vegetable oils (i.e.. 

The crystallization of the minor component in incompatible polymer blends starts sometimes at distinctly larger undercoolings than in the pure polymer, and proceeds in several separated steps. After a short survey on the history of the effect in the available literature, the several types and the origin of this "fractionated crystallization" as observed in some selected systems are described. The information on the blend which can be deduced from the effect is discussed, and the consequences for the blend processing and properties are investigated. 

This survey on the literature indicates that only few data are available on the droplet crystallization phenomena in incompatible polymer blends. Moreover, these observations are partly not completely explained, and, where explained, these explanations are partly not satisfying or contradictory. In the next chapter, therefore, experimental results for some selected systems as investigated by the authors are presented with the aim to show all faces and properties of fractionated crystallization in detail, and to contribute to a better understanding of the origin of the effect. 

It should be pointed out that there is no direct physical relation between the phenomenon of fractionated crystallization and the number and the size of spherulites in the pure polymer. Whereas the occurrence of fractionated crystallization is related to the ratio between the number densities of dispersed polymer particles and primary nuclei, the size and the number of spherulites are additionally influenced by the cooling rate and the crystallization temperature. There is, therefore, also no relation between the fractionated crystallization and the type of the arising crystalline entities (complete spherulites, stacks of lamellae,...) both in the pure and in the blended material. There is, finally, no relation between the scale of dispersion which is necessary for the occurrence of fractionated crystallization and the spherulite size in the unblended polymer. 

A completely different behavior is reported for blends in which the crystallizable phase is dispersed. Fractionated crystallization of the dispersed droplets, associated with different degrees of undercooling and types of nuclei is the rule. The most important reason is a lack of primary heterogeneous nuclei within each crystallizable droplet. An important consequence of fractionated crystallization may be a drastic reduction in the degree of crystallinity. 

However, for polymer blends in which the crystallizable phase is dispersed into fine droplets in the matrix, crystallization upon cooling from the melt can sometimes occur in several steps (fractionated crystallization) that are initiated at different undercooling, often ending up with a crystallization at the homogeneous crystallization temperature T, [Aref-Azar et al., 1980 Bailtoul et al., 1981 Ghijsels et al., 1982 Santana and Muller, 1994]. 

From the fractionated crystallization behavior and the blend morphology, one can determine the... 

The thermal behavior of PS/LDPE blends has been investigated by Baitoul et al. [1981]. A clear indication of the fractionated crystallization was deduced from the appearance of two additional crystallization peaks around 71 and 64°C in all blends in which LDPE was the dispersed phase. Eurthermore, the crystalhzation kinetics was found to slow down severely when the content of PS was raised. 

Zhou and Hay [1993] investigated the crystallization in LLDPE/PP blends. They reported that the extent of crystallization in PP droplets is seriously hindered by the low nucleation density of PP, resulting in a serious drop of the degree of crystallinity during the isothermal measurements. From these experiments it could be predicted that cooling from the melt would result in a fractionated crystallization (30 wt% PP) or even homogeneous crystallization (10 wt% PP). Similar results had already been reported by Long et al. 

Klemmer and Jungnickel [1984] have reported on the fractionated crystallization of POM in an HOPE matrix. They found an additional crystallization peak of POM to occur 14°C lower than the bulk crystallization peak. This was attributed to the fractionated crystalhzation of POM, caused by an interface-induced additional inhomogeneous nucleation and crystallization. It was shown that this phenomenon only occurs in those blends where the number of the dispersed particles was higher than the number of available heterogeneous particles. Moreover, the preparation method clearly influenced the fractionation due to the change of the particle sizes - fractionated crystallization has been observed only in melt-mixed blends. 

Polyethylene as Dispersed Phase. PP/PE blends have been studied extensively by several authors. Zhou and Hay [1993] reported that the dispersed LLDPE droplets in PP/LLDPE blends showed problems in nucleating at the normally expected bulk crystallization temperature, r. Also, a serious decrease of the degree of crystallinity from isothermal measurements, as the LLDPE content decreased, could be observed. Contrary to these observations, Muller et al. [1995] stated that the LLDPE droplets do not exhibit fractionated crystallization when they are dispersed in a PP matrix (although they do in a PS matrix), because of the nucleating effect of the solidified PP matrix on the LLDPE droplets. 

Figure 3.47. Retarded and/or fractionated crystallization causing coincident crystallization in PVDF/PA-6 and PVDF/PBT blends. Influence of the blend composition (a) and the number of extrusion cycles Z (b) [Frensch and Jungnickel, 1989].

Generally, the melting point of the two forms decreased with the increase in the HOCP fraction, for blends prepared with a constant content of PB (Table 6.14). The tendency of PB to crystallize in form 1 preferably than in form 11 resulted to be strongly influenced by blend composition. The melting point of PP slightly decreased with the decrease in its fraction in the blend or with the increase in the HOCP content. 

The values offy reported in Table 6.17 show that the crystallization in form I is the favorite for blends prepared with small fractions of PB-1 by increasing the PB-1 content in the blends, the fraction crystallized in form II increased. These results are in good agreement with the trend observed for the thermal data obtained by DSC. PB-l/HOCP binary blends containing up to 30% of HOCP showed that PB-1... 

---