Exploring the polymorphic behavior of a β-nucleated propylene-ethylene random copolymer under shear flow

2.1 Materials

P-E random copolymer (prepared via Ziegler-Natta catalyst) with a melt flow rate of 1.8 g/10 min (230°C/2.16 kg) was supplied by SINOPEC Beijing Yanshan Company. The weight-average molecular weight (M_w) is 5.77 × 10⁵ g mol^–1, and the number-average molecular weight (M_n) is 1.63 × 10⁵ g mol^–1. The content of ethylene units is ~5.6 mol%, as measured by ¹³C-NMR. Calcium pimelate (Ca-Pim), an efficient β-NA for P-E random copolymer,33, 37, 38 was synthesized by neutralization reaction between Ca(OH)₂ and pimelate acid.

2.2 Sample preparation

The P-E random copolymer, Ca-Pim, and an antioxidant (Irganox B215, Ciba-Geigy) powders were melt mixed in an internal mixer (HAAKE Rheomix OS) at 200°C for 6 min. The weight fraction of Ca-Pim and antioxidant were both 0.1 wt%. The obtained mixture was compressed into films of 200 ~ 300 μm thick at 200°C. Samples in the shape of a ring (i.d. = 10 mm, o.d. = 20 mm) were cut from the pressed films for the WAXD measurements. The melting temperatures (T_m) of the β and α crystals in the β-nucleated P-E random copolymer are 129.9°C and 146.4°C, respectively.

2.3 In situ WAXD measurement

The in situ WAXD measurement was performed at the beamline 1W2A in the Beijing Synchrotron Radiation Facility (BSRF). All diffraction patterns were captured by a MAR CCD (MAR-USA) detector with a resolution of 2048 × 2048 pixels (pixel size: 79 × 79 μm²). The two-dimensional (2D) WAXD patterns were converted to one-dimensional (1D) intensity profiles after background correction by a home-written code using Python.

The overall crystallinity index (X_c) of the β-nucleated P-E random copolymer was calculated by the following equation:

(1)

where the ∑A_crystalis the sum of crystalline peak areas and the A_total is the direct integration of the whole intensity profile.

An example of calculation for ∑A_crystal and the areas of the diffraction peaks of α, β, and γ crystals are displayed in Figure 1. The peaks are indexed according to the previous literature.39, 40 The relative fraction of β crystals (k_β) and the fraction of γ crystals (k_γ)41 can be expressed as follows:

(2)

(3)

Correspondingly, the fraction of β, γ, and α crystals can be expressed as follows:

(4)

(5)

(6)

A Linkam CSS-450 high-temperature shearing stage (Linkam Scientific Instruments, Ltd.) was used to control the shear flow and temperature. The samples were positioned between two metal plates coated by Kapton films. A hole was left for the X-ray to pass through. The hole has a diameter of 2.8 mm, located at a radius of 7.5 mm to the center of the rotation axis. Schematics of the temperature and shear protocols are shown in Figure 2: (A) heated the sample to 200°C at 30°C min⁻¹ and held for 3 min and (B) cooled down to 125°C at 30°C min⁻¹. A steady shear with rate = 5 s⁻¹ was immediately applied to the sample with different shear time (0, 2, 4, 6, 10, and 20 s) as soon as the temperature reached 125°C. Correspondingly, the shear strains were 0, 10, 20, 30, 50, and 100, respectively. The relatively weak shear rate was chosen to avoid melt fracture considering the high viscosity of the polymer used in this study.

2.4 Scanning electron microscopy

The etched samples were observed on a scanning electron microscopy (SEM) (JSM-6700F, JEOL), operated with an accelerating voltage of 5 kV. All the samples were sputter-coated with platinum to avoid charging. The selective permanganic etching method proposed by Olley and Bassett42 was used to show the crystalline morphology. The samples were crystallized at 125°C for 40 min, then cooled to room temperature at a rate of 30°C min^–1. The etching solution was prepared by dissolving 25 mg potassium permanganate (KMnO₄) in 50 mL concentrated orthophosphoric acid. After etching for 70 min, the samples were sequentially washed with a mixture of concentrated sulfuric acid and distilled water (volume ratio = 2:7), hydrogen peroxide, and acetone.

3.1 2D WAXD results

The selected 2D WAXD patterns of the β-nucleated P-E random copolymer isothermally crystallized at 125°C under quiescent and shear conditions are shown in Figure 3. The patterns in the first row exhibit a diffuse halo, corresponding to the amorphous polymer melt. With isothermal time increasing, two isotropic reflections appear in the quiescent polymer which can be indexed as the (110)_β and (111)_β of β crystals. Upon shearing, the reflections appear earlier as compared to the quiescent condition. The intensity of (110)_β decreases with shear strain. When the shear strain exceeds 50, the reflections of β crystals are rather weak. Meanwhile, signals from α and γ crystals are more obvious. These reflections of α and γ crystals are slightly anisotropic, indicating a certain degree of orientation. The diffraction features are discussed in more detail below.

3.2 Crystal orientation

The orientation behaviors of different crystal forms can be expressed by the azimuthal intensity distributions of characteristic diffraction rings, as shown in Figure 4. The (110)_β is always uniform under different shear strains, indicating that the orientation of β crystals is independent of shear. Li et al.43 found that relaxation of oriented molecular chains was beneficial to the formation of β crystals and the oriented β lamellae can be obtained by epitaxial growth on the needle-like β-NA. In our case, the Ca-Pim particles are nearly spherical, thus the β crystals are always isotropic under shear flow. The (110)_α/(111)_γ reflection becomes more anisotropic with increasing shear strain. This indicates that the oriented chains induced by shear flow tend to form oriented α nuclei.22, 23 Subsequently, it is probable that the γ crystals epitaxially grow onto the oriented α crystallites.44-46 According to the literature,47, 48 the c-axis in mother lamellae of the α crystals is parallel to the shear direction, while the daughter lamellae keep a constant 80° angle with the mother lamellae. As shown in Figure 4, the (110)_α/(111)_γ reflection has an arc at 0° indicating that the α and γ crystals exhibit a certain degree of orientation upon shear.

3.3 Crystallinity and polymorphism

To characterize the evolution of different crystal forms in the β-nucleated P-E random copolymer, the 1D WAXD intensity profiles are shown in Figure 5. Under the quiescent condition, the β crystals and α crystals appear almost simultaneously. When the shear strain = 10, 30, and 50, the (110)_β reflection appears later and its intensity becomes weaker as compared to the quiescent condition. When the shear strain further increased to 100, no β crystals can be observed. Hence, the formation of β crystals is suppressed by shear flow. On the other hand, the induction time of α and γ crystals decreases from 450 to 120 s as the shear strain increased from 0 to 100.

Figure 6A presents the variation of the overall crystallinity index (X_c) of the β-nucleated P-E random copolymer during isothermal crystallization under the quiescent and different shear conditions. It can be seen that both the crystallinity and the crystallization kinetics are affected by the shear flow. Since the crystallinity index does not go flat at the tail, the crystallization profiles can only be described by considering both primary and secondary crystallization process. The overall crystallization kinetics can be fitted by the following equation49:

(7)

where, X_t and X_p,∞ are the crystallinity at time t, and at the end of the primary process. X_p,∞ manifests itself as a sudden slowdown of X_t. Z_p is a rate constant involving primary nucleation and lamellar growth. The exponent n is a constant depending on the nucleation and growth mechanism. And k_s is the rate constant for secondary crystallization. The parameters obtained from the fitting are listed in Table S1-S4 in the supporting information. Accordingly, the half-time of primary crystallization (t_0.5) can be calculated as follows:

(8)

Under the quiescent condition, the final X_c is ~18.2% and the t_0.5 is ~950 s. With a mild shear (shear strain = 10 and 20), the X_cs are a little higher than the static condition and the t_0.5 decreases to ~800 s. When a relatively intense shear (shear strain = 100) is applied to the β-nucleated P-E random copolymer, the final X_c increases to 20.8%, and the t_0.5 further decreases to ~250 s. The enhancement of the crystallization rate of polymers by shear flow has been known. However, the X_c is generally not affected by shear in iPP homopolymer and its mixture with β-NA.3, 50, 51 Enhanced crystallinity was observed in a propylene-1-butylene random copolymer under shear condition.3 This may be a general feature for random copolymers.

The fractions of the α, β, and γ crystals as a function of time are shown in Figure 7A-C. The formation of α and γ crystals is enhanced both in the overall content and the kinetics. On the other hand, the formation of β crystals is greatly hindered. Remarkably, when the shear strain reaches 100, the β crystals vanish completely, which is never seen in previous studies on β-nucleated iPP under shear. As shown in Figure 7D, under the quiescent condition, the value of X_β is ~7.5%. When shear strain = 20, the X_β decreases to 3.5%. As shear strain further increased to 100, the value of X_β is nearly 0, while the X_α and X_γ reach the highest values. Under shear conditions, the γ crystals become predominate in the β-nucleated P-E random copolymer.

Figure 8A displays the t_0.5 of the α, β, and γ crystals as a function of shear strain. The t_0.5 of the β crystals is always longer than that of the α and γ crystals. On the other hand, the t_0.5 of the α and γ crystals are comparable with each other and both decrease with shear strain. Thus, it is probable that the kinetics of γ crystals is similar to that of α crystals. To further reveal the relationship between α and γ crystals, the k_γ is plotted as a function of time. As shown in Figure 8B, it is interesting to find the k_γ is almost independent of the shear strain and crystallization time. These results provide strong evidence of the correlation between α and γ crystals and support the existence of epitaxy.

3.4 Crystalline morphology

The effects of shear flow on the polymorphism can also be confirmed by morphology. It is known that β crystals exhibit a bundled structure,25, 37, 52 while a cross-hatched structure forms in α and γ crystals.53, 54 Unfortunately, the epitaxial relationship between α and γ crystals makes it difficult to distinguish them by morphology.55 As shown in Figure 9A1-A3, the bundled-like β crystallites predominate under quiescent crystallization, and its content decreases with the shear strain. On the other hand, the fractions of cross-hatched lamellae increase with the shear strain (shown in Figure 9B,C). It should be mentioned that a small number of bundled-like structures (β crystals) are also seen in Figure 9C which is probably formed during cooling instead of during isothermal crystallization. These results are in line with the WAXD results.

3.5 Mechanism of the suppression of β crystals by shear flow

The suppression effect of shear flow on the formation of β crystals in the β-nucleated iPP samples has been reported.30, 31, 56 However, the combination of shear-induced β nuclei and β-NA induced nucleation makes it difficult to reveal the effects of shear flow on the nucleation and growth of β crystals. On the other hand, the shear-induced β nucleation was eliminated in the β-nucleated P-E random copolymer (Supporting Information, Figure S1), which makes the situation simpler.

Based on our results, a thermodynamic illustration is drawn in Figure 10. According to the classical nucleation theory, the barrier for homogeneous nucleation can be expressed as57:

(9)

where σ, σ_e are the surface free energies of the side and the end surface of the nucleus with a quadratic cross-section, T_m, ΔH_m, ΔT, and ρ_c are the equilibrium melting temperature, the enthalpy of fusion per unit volume, the degree of supercooling, and the crystal density, respectively.

As shown in Figure 10A, under the quiescent condition, the free energy barrier for the β nucleus (ΔG_β*) is higher than that of the α nucleus (ΔG_a*). Therefore, no β crystals are formed in the P-E random copolymer (Supporting information Figure S1). By adding β-NA, the free energy barrier of the β nucleus (ΔG*_{β, N.A}) is dramatically decreased because of the lower surface free energy (σ or σ_e) in proximity to the surface of the β-NA. The ΔG*_{β, N.A} should be comparable to ΔG_a*, which can be inferred from the fact that both the α and β crystals are formed in a large number in the sample.

As shown in Figure 10B, under the shear condition, the free energy of the oriented melt increased by T*ΔS because of the decreased conformational entropy of polymer melt. Therefore, the nucleation barriers for the two crystal modifications (ΔG*_{α, shear} and ΔG*_{β, shear}) decrease simultaneously. Following this concept, in principle, both the nucleation of the α and β crystals will be enhanced if the crystallization occurs from the same melt state. Now the key question is: whether the effect of shear and β-NA can add up to promote the crystallization of the β crystals. Experimentally, the answer is no. Whether their effects can be added up depends on the influence of the Ca-Pim particle on the melt flow. The Ca-Pim particles are of micrometer size. According to the literature,58 the existence of the particles will hinder the chain alignment under flow. For poly(lactic acid) and iPP, inorganic nucleating agents have no effect on the strain-induced crystallization when the strain rate is sufficiently high.59, 60 If the two paths could not add up, suppose that the ΔG*_{β, shear} > ΔG*_{β, N.A.}, the lowest nucleation barrier for the β crystals is still ΔG*_{β, N.A.,} while the ΔG*_{α, shear} decreases dramatically with the shear strain. Under this condition, the nucleation of β crystals will be greatly suppressed.

The above illustration agrees with experiments very well. As shown in Figure 7B, the induction time of the β crystals is basically unaffected by the shear strain. On the other hand, the crystallization kinetics of the α and γ crystals is significantly accelerated by shear. Under intense shear, before the formation of β crystals, the polymer melt is already consumed by the α and γ crystals. Therefore, the polymorphism of the P-E copolymers is effectively controlled by the nucleation kinetics. The γ crystals closely related to the α crystals are not shown in Figure 10. It is known that a decrease of the regular isotactic sequences, the introduction of comonomers and isothermal crystallization are favorable to the formation of γ crystals.61-64 It is possible that shear flow makes more chain segments containing ethylene comonomers crystallize into the γ crystals. This explains why the fraction of γ crystals and the overall crystallinity increase with the shear strain.