Toughening mechanism in impact polypropylene copolymer containing a β-nucleating agent

Abstract

In this work, the effects of β-nucleating agent and annealing treatment on the crystallization and amorphous phase in impact polypropylene copolymer (IPC) were profoundly displayed to reveal the toughening mechanism of the modified product. The toughness of β-nucleated IPC was improved dramatically compared with samples without nucleation agent. The chain mobility measured by dynamic mechanical analysis (DMA) in the rigid amorphous fraction (RAF) weakened with increasing annealing temperature, while that in the mobile amorphous fraction (MAF) strengthened under low annealing temperatures (not more than 124 °C) and then weakened at higher temperatures. On the one hand, the impact energy is effectively dissipated by slipping of loose β lamellae. On the other hand, the voids in loose β crystals multiply shear yielding. By contrast, dense α crystals in neat IPCs provide fewer voids and the major pathway of energy dissipation is crazing induced by soft dispersed particles. After annealing treatment, all the β-nucleated IPCs became brittle, which was mainly ascribed to the reduction of β crystals through β–α transformation.

---

1. Introduction

Impact polypropylene copolymer (IPC), one of the most important multiphase polymer systems synthesized via continuous in situ copolymerization with propylene and ethylene in a gas-phase reactor, consists of a homopolymer polypropylene (HPP) matrix, an amorphous ethylene propylene random copolymer (EPR) and an ethylene–propylene block copolymer with different segment lengths (EbP).^1–5 It has been widely used in the automotive industry, in appliances and other durable goods applications for its excellent comprehensive performance, especially for its good toughness.^2,6 The outstanding toughness is attributed to the special core–shell structure of dispersed particles.^7–9 The fracture resistance would be contributed by cavitation in soft dispersed particles and crazing in matrix, with the outer shell improving interfacial adhesion between HPP matrix and particles.¹⁰ However, on the one hand, the core–shell structure of dispersed particles may become incompleted or even be destroyed during processing. On the other hand, high content of dispersed particles is not allowed as it will cause big loss in strength. There's no guarantee that those dispersed particles could work effectively. To get more superior materials and extend its application range, re-modification of IPC by other toughening methods is also meaningful and necessary. Most important of all, the adjustment of crystalline polymorphism and amorphous fraction in HPP matrix may be a potential pathway and it plays a role equally crucial to the soft dispersed particles. Using β-nucleating agent (β-NA) and annealing treatment both have been proved to be effective ways to tailor them through controlling crystallization behavior and chain mobility of HPP matrix.^11–13

Generally, the HPP can crystallize in at least three forms: stable monoclinic α crystal, metastable trigonal β crystal and orthorhombic γ crystal in most stable form.^14,15 Among them, the α form occurs mostly till part of them is replaced by loose β crystal with the introduction of β-NA. The impact strength of β form is better than that of α crystal and it may be attributed to a stress-induced transformation of less dense β crystal, which allows slipping and deformation of lamellae as well as β–α transition to enhance the energy dissipating.^15,16 In addition to the crystallization of HPP matrix, the amorphous interlayer connecting lamellae is also affected by adding β-NA. The motion of segments here are extremely relevant to toughness as they play an important role in transferring impact force through the whole sample. The impact energy was dissipated partly by viscoelastic relaxation of HPP amorphous domain. Previous results have suggested that the motion of segments is determined by the density of chain entanglement and segmental number in the amorphous region, which could be characterized by their relaxation behaviors.¹⁷ A linear increasing relation between intensity of relaxation and impact strength have ever been revealed.^11,18

Processing condition is another key factor in crystallization. Usually, annealing treatment at temperature located between glass transition temperature (T_g) and melting temperature (T_m) is necessary after a rapid cooling and solidifying course. The thermal annealing may induce many changes, including rearrangement of molecular chains, perfection of defective crystals and chain mobility of the amorphous fraction (MAF) and rigid amorphous fraction (RAF).^17,19,20 Some works have proved that the toughness of polymeric materials can be largely improved at certain annealing temperatures (between 120 and 130 °C for PP).^12,17,21 They thought that the decreased number of chain segments in the amorphous region after recrystallization leads to the formation of microvoids, which promote the lamellae to slip or elongate along the impact direction and induce the intense plastic deformation during the fracture process. But in other cases, rubber particles modified copolymers embrittle on annealing because the voids are hardly observed after that.²² So far, no uniform conclusion has been drawn.

Thus in this work, the synergetic effects of β-NA and annealing on crystallization behavior, segment mobility in amorphous phase and impact strength of IPC are investigated to establish structure–property relation. By comparing those behaviors of neat IPC and β-nucleated IPC under different annealing treatment, the toughening mechanism in IPC system is discussed and systematically.

2. Experimental

2.1 Materials

The IPC (SP179) is a commercial material purchased from Sinopec Qilu Petrochemical Co. of China (the structural informations of IPC can refer to ESI†), with M_n of 4.39 × 10⁴, M_w of 1.74 × 10⁵ and a density of 0.91 g cm⁻³. The β-NA is a powder of metal salt (calcium tetrahydrophthalate, tradename NAB83), supplied by GCH technology Co., Ltd.²³ To avoid oxidative degradation, a special type of antioxidant (Irganox 1010) is adopted from Virtulla Tianjin Technology Co., Ltd.

2.2 Sample preparation

A master-batch containing 1 wt% β-NA were first prepared by blending IPC and β-NA on a torque rheometer (XSS-300, KCCK, Shanghai, China) under 180 °C for 8 min with a screw speed of 60 rpm (a small amount of antioxidant was also added). Then the master-batch and pure IPC were further melt blended and chipped by co-rotating twin screw extruder (PRISM TSE 16 TC, Thermo Scientific, Waltham, UK), at a screw speed of 20 rpm and barrel temperatures set as 160, 180, 180, 180 °C from hopper to die successively. Finally, the IPC materials with 0, 0.05 wt% and 0.1 wt% β-NA were obtained. Standard specimens of DMA (17.5 × 12 × 3 mm³) and impact strength (80 × 10 × 3.4 mm³) measurements were hot pressed at plate vulcanizing machine under 180 °C for 8 min. After that, they were annealed in oven at 120–150 °C for 30 min separately.

2.3 Differential scanning calorimeter (DSC)

All thermal behaviors were measured in nitrogen atmosphere with a DSC (Q100, TA Instruments Corporation). A heating rate of 10 °C min⁻¹ was adopted to record melting traces. In order to maximize the signal, the weight of each sample was kept at 8 + 0.5 mg. Multiple melting peaks were separated by Gauss-fitting method to obtain melting fusion of α crystal and β crystal so that their crystallinities could be calculated. The standard heat fusion of α and β crystals of PP in IPC are 178 and 170 J g⁻¹ respectively.^24,25

2.4 Wide angle X-ray diffraction (WAXD)

The film samples were prepared by compression molding and WAXD patterns of them were obtained on a D/Max-2550PC X-ray diffractometer with the Cu K_α radiation (λ = 1.540 Å) at room temperature. The continuous scanning angle (2θ) ranged from 3° to 40°, at a voltage of 40 kV and a current of 250 mA. The d-spacing of the sample was scanned at 5° min⁻¹, with a step of 0.02°. The crystallinity (X_c) was calculated according to the following equation:^26,27

After the amorphous background has been extracted, A_crys and A_amorp are the areas of the fitted crystal peaks and amorphous contribution peak respectively.

2.5 Small angle X-ray scattering (SAXS)

SAXS experiments were performed at XEUSS SAXS/WAXS SYSTEM (French, XENOCS SA) with a X-ray generator (30 kV and 650 mA), which provides Cu K_α radiation (the wavelength was 1.5411 Å). The distance between sample and detector was 2522 mm. The SAXS signal was collected with a two-dimensional gas-filled wire detector and each SAXS pattern was collected within 600 s. The measured SAXS intensity was calibrated for background scattering by direct model fitting, and then the one-dimensional correlation function k(z) is calculated by cosine Fourier transformation, seen in eqn (2).

2.6 Dynamic mechanical analysis (DMA)

The chain mobility in amorphous region was measured using a DMA (TA Instruments Corporation, USA, Q800) in single cantilever mode. Multi-frequency–strain measurements were performed under a heating rate of 3 °C min⁻¹ from −70 °C to 150 °C and at frequencies of 1 and 10 Hz respectively. Given that a similar trend was observed under two frequencies, curves at frequency of 1 Hz were exhibited in this work.

2.7 Notched Charpy impact strength measurement

The notched Charpy impact strength of specimens was tested on a notch impact instrument (MTS systems Co. Ltd., China) according to GB/T 1043.1-2008. V-shaped notches about 45° were cut (2 mm depth) by using a semi-automatic notcher before measurement. Two days later, the specimens were kept in a temperature & humidity chamber at 0 °C and 23 °C for 12 h respectively. The measured values were obtained by averaging eight specimens.

2.8 Scanning electron microscopy (SEM)

The standard specimens after impact strength measurement were collected. Impact fracture section in one half was etched in xylene for 2 h at 50 °C to remove the amorphous EPR component while the other half wasn't. Morphology of fractured surface of samples was observed by SEM (Hitachi, S-4800) after sputter-coated with gold powder.

3. Results and discussion

3.1 Effects on crystalline phase

As well known, both introduction of β-NA and annealing treatment are common methods to tailor crystal content as well as crystalline structure. In order to investigate the influence of β-NA and annealing treatment on crystallization of IPC, WAXD and DSC techniques were used to investigate the pure and modified IPC samples containing 0.05 wt% and 0.1 wt% β-NA. As seen in Fig. 1(a), there are six characteristic diffraction peaks at 2θ angles of 14.0°, 16.0°, 16.8°, 18.6°, 21.2° and 21.9°, corresponding to the (110), (300), (040), (130), (111) and (131 & 1041) lattice planes of PP. The (300) lattice plane indicates the existence of β-PP crystals while the others are diagnostic diffraction peaks of α-PP. A weak peak at (300) lattice plane is observed in neat IPC, indicating the formation of a small quantity of β-PP crystals induced by rapid molding.²⁸ Obviously, peak at (300) lattice plane is greatly strengthened after introducing β-NA. In addition, two distinct melting peaks for β-nucleated IPC appear within the temperature ranges from 140 to 170 °C while only a single peak exists in neat IPC, as seen in Fig. 1(b). It can be documented that the lower one in double melting endotherms owed to melting of β crystal (about 149 °C) while the other at higher temperature corresponded to α crystal (about 167 °C), in accordance with the literatures.^15,29,30 Compared with the neat IPC, the melting curves of β-nucleated IPC indicate that the β-NA can effectively induce β crystal in IPC sample. The analogous results have also been reported.^23,31,32 Xu et al.³¹ found that this calcium tetrahydrophthalate was a high efficient β-NA for the IPC (tradename J340), where the relative fraction of β-crystal could reach as high as 93.5% with only 0.03% β-NA. Luyt et al.³² believed the efficacy of the β-NA was dependent on the chemical composition of IPCs, more specifically on the sequence length of crystallisable propylene units. As to the minor PE crystals, stemmed from the long ethylene sequences in ethylene–propylene block copolymer, there is little change on them for IPC samples with different contents of β-NA (seen the melting peaks at 110 °C in Fig. 1(b)) and its effect on properties could be ignored in the discussion below.

Fig. 1 WAXD profiles (a) and DSC melting curves (b) of IPC samples with different contents of β-NA.

The effect of annealing on the crystalline microstructure has been reported extensively in previous works and it is thought to be an efficient way to achieve the thermodynamic equilibrium state in a short time and improve properties in a certain extent.^11,12,17 On the one hand, it may bring changes on crystallinity and crystal form. For neat IPC, there is little change for all diffraction peaks but a remarkable decrease of the diffuse scattering peak (the part surrounded by dotted lines in Fig. 2(a)) after annealing, indicating some chains in amorphous phase folding into crystals. As to β-nucleated IPC, its annealing treatment was conducted at 124 °C as it was reported to be the most suitable temperature for formation of β-PP crystals.¹⁵ It can be seen from Fig. 2(b) that the characteristic diffraction peaks here remained unchanged after annealing, suggesting little change on crystal form. The similar results of β-nucleated IPC appear after being annealed at different temperatures (23–140 °C) except at 150 °C. A distinct rise of (110) lattice plane is observed when annealing at a high temperature of 150 °C, which is beneficial to form α-PP crystals. The related data of crystallinity are calculated by WAXD and DSC separately and the results are listed in Table 1 (take the β nucleated IPC for an example). Though the values of crystallinity obtained by two methods are different, the changing trends are accordant. After annealing treatment, the content of β crystal decreases, and both crystallinities of α crystal and total crystallinity increase with elevated temperatures for β-nucleated IPC. When the annealing temperature is above 140 °C, the changes are the most notable. Obviously, those should be ascribed to β–α transformation of crystals during annealing. On the other hand, annealing may bring changes on lamellar thickness. Fig. 3 gives the melting curves of β-nucleated IPC subject to different annealing treatments. The melting temperatures are proportional to lamellar thickness according to the Gibbs–Thomson equation.^33,34 Generally, the detected crystals would recrystallize and form into more stable and thicker crystals undergoing annealing. When the samples are annealed at 124 °C for different time, the melting peaks at higher temperatures (α-PP crystals) almost keep constant but those at lower temperatures (major β-PP crystals) show a little decrease compared with unannealed one. The former indicates that the annealing time has little effect on lamellar thickness of α-PP crystals, but the latter seems abnormal. Actually, the lower peaks not only come from β-PP crystals but also are from partial α-PP crystals, due to wide molecular weight distribution of IPC. Though the annealing temperature of 124 °C facilitates generating β nucleus, new crystals during recrystallization are still the α-form primarily as they are born basing on pre-existing lamellae. Thus, the mixing information of the lower temperature peak can be separated. It is thought that the melting points of β-PP crystals still stays at about 150 °C and the enlarged melting enthalpy at higher temperatures should be ascribed to new α-PP crystals induced by recrystallization. After being annealed at different temperatures, melting peaks of both crystals may shift to high temperatures remarkably, indicating that both kinds of lamellar crystals thicken. Through recrystallization, original imperfect crystal could develop to more perfect form.

Fig. 2 WAXD profiles of neat IPC (a) and β-nucleated IPC (0.1 wt%) samples after annealing at 124 °C for different time (b) and at different temperatures (c).

Table 1 Crystallinity measured by WAXD and DSC for β-nucleated IPC (0.1 wt%)

Annealing treatment	WAXD			DSC
	Xβ	Xα	Xt	Xβ	Xα	Xt
Unannealed	0.170	0.157	0.327	0.276	0.107	0.383
124 °C, 30 min	0.154	0.187	0.341	0.240	0.147	0.387
130 °C, 30 min	0.133	0.231	0.364	0.198	0.184	0.382
140 °C, 30 min	0.126	0.256	0.382	0.171	0.226	0.397
150 °C, 30 min	0.108	0.320	0.428	—	—	∼0.433

---

Fig. 3 Endothermic curves of 0.1 wt% β-nucleated IPC after annealing at 124 °C for different time (a) and at different temperatures for 30 min (b).

3.2 Effects on amorphous phase

In addition to crystallization, the structure and relaxation behavior of amorphous phase is another important factor determining impact property.^11,17 The amorphous dispersed phase in IPC consisted of ethylene propylene copolymer which would promote energy dissipation by deformation and crack initiation, is responsible for the major toughening effect.³⁵ Furthermore, the chain mobility of amorphous HPP matrix located at inter-spherulite and intra-spherulite, determines the further development of crack.³⁶ The relaxation behavior, called as chain mobility too, is usually characterized by DMA. Generally, either lower peak temperature or greater intensity of tan δ indicates the enhancement of chain mobility in amorphous region. Quantitative or at least qualitative analysis about relaxation strength and fracture resistance relationship has been found that the impact strength increases linearly with the strengths of chain mobility.^11,18

In the case of IPC, three loss peaks turn up from about −60 to 140 °C (Fig. 4). Our previous work² has proved that the middle peak at about 25 °C belongs to glass transition of HPP (β relaxation) and that at −30 °C corresponds to glass transition of EPR phase in IPC. As to the loss peak located at about 90 °C (called α relaxation), though there are many related researches reported, it is a controversial topic till now. There is no doubt that it is induced by HPP fraction but it is quite different from the β relaxation of HPP which is related with glass transition.³⁷ Up to date, there are two main viewpoints about the origin of α relaxation peak: (a) the diffusion of rigid amorphous fraction (RAF) which is strongly restricted by crystalline phase; (b) the relaxation resulting from pre-melting of defective crystals.³⁸ Both amorphous and crystalline phases are therefore thought to be affected by this transition. It can be clearly seen in Fig. 4 that the temperature of α relaxation peak (α₁, α₁′ and α₁′′) falls and the intensity increases as the content of β-NA in IPC rises. This could be ascribed to the increased RAF and imperfect β-crystals with increasing β-NA. The temperatures of β-relaxation peaks almost keep identical but the intensity increases, indicating enhancement in mobility and quantity for HPP amorphous phase. Furthermore, the neat IPC presents nearly the similar relaxation peak accounting for the glass transition of EPR compared with β-nucleated IPCs, indicating that the content of β-NA hardly affects EPR amorphous phase.

Fig. 4 Mechanical loss factor (tan δ) versus temperature of unannealed IPC with different β-NA contents.

It is reported that annealing has a positive influence on toughness when it significantly changes the relaxation behavior of HPP molecular chains.^11,17 So in this work, the peaks of tan δ for different β-nucleated IPCs before and after being annealed were studied in detail. As shown in Fig. 5, either increasing annealing time or increasing annealing temperature can lead to the shifting toward the higher temperature of α relaxation peak. Chain motions in the RAF become slow. It may be ascribed to denser chain packing so that interchain interactions increase during the crystal thickening and β–α crystalline transformation. The similar phenomenon has been reported in the work of Hedesiu et al.²⁰ They believe that annealing at temperatures above 110 °C causes increases in the lamellar thickness and a decrease in the chain mobility of semi-rigid fractions (the same to RAF) for iPP. Thus, it is reasonably thought that annealing can also lead to slow the chain mobility for HPP RAF in IPC. Meanwhile, the peak temperature of β-relaxation doesn't show an obvious dependence of annealing temperature and they are almost identical. But compared with unannealed IPC, the intensity of β peak increases first when the temperature is not more than 124 °C, and then it turns down at higher annealing temperatures. That is to say, chain mobility of mobile amorphous fraction (MAF) is enhanced by annealing but it is depressed at higher annealing temperature. Fu et al.¹² thought that the increased chain mobility should be due to the decrease in density of MAF as amorphous phase becomes looser by part of them turning into RAF. However, the segment amount in this region decreases and the amorphous phase will be restricted by increasing RAF at higher annealing temperatures.

Fig. 5 Mechanical loss factor (tan δ) versus temperature of annealed IPC with different β-NA contents: (a) 0.05 wt%, (b) 0.1 wt%.

3.3 Morphological parameters measured by SAXS

The effects of β-NA and annealing on microstructure can be revealed by SAXS. In a non-ideal two-phase system, L is assumed to be the sum of average thicknesses of lamellae (L_c), amorphous layers (L_a) and transition zones (L_tr) (L = L_c + L_a + 2L_tr).³⁹ Compared with the relaxed MAF, the transition zone is termed to the ordered RAF, distributed at the interface between the crystal and mobile amorphous layers. Those morphological parameters of the crystalline and amorphous fraction can be revealed by the correlation function k(z) (the eqn (2) in the Experimental section). The approximate distance between the crystalline lamellae can be obtained from the position of the maxima in the SAXS correlation function curves that describe the distribution of the lamellae and amorphous thickness. The distance z in real space at the first maximum of k(z) curve corresponds to the long period L and that of the intersection point represents the average thicknesses of lamellae L_c (Fig. 6).^40–42 To simplify the analysis, the layers of MAF and RAF could be considered as a whole, that is L_A = L_a + 2L_tr. The results show that all long periods of neat IPC and β-nucleated IPC increase with the increasing annealing temperature. The latter are over 18 nm which is greater than that in IPC (about 12 nm), resulting from loose structure of β crystals. The L_c of IPC also increases resulting from lamellar thickening after annealing. However, it is difficult to distinguish the changes in β-nucleated IPC as the lamellae here are the mixture of α-PP and β-PP crystals. The small peaks indicated by the arrows in Fig. 6(b) just verify that there are more than one periodic layer structure. Thus, we obtained information about L_c by melting points based on the relationship between lamellar thickness and melting temperature.^34,43 The results in Fig. 3 suggest that the L_c of β-nucleated IPC increases with elevated annealing temperature. Meanwhile, as indicated by DMA results, the chain mobility slows further with the increase of annealing temperature, indicating that the thickness of amorphous layer decreases. Hence, the increase in L here is mainly ascribed to the thickening thickness of lamellae. Consequently, the results of SAXS analysis about the effect of β-NA and annealing treatment on IPC are consistent with conclusions discussed above in this work.

Fig. 6 Curves of one-dimensional correlation function k(z) for IPC (a) and β-nucleated IPC (b) after annealing at different temperatures.

3.4 Relationship between impact property and microstructure

It is well known that the properties of the final products are closely related to the microstructure, including crystalline phase, amorphous phase as well as multiphase morphology. Here, the introduction of β-NA and annealing treatment in IPC have brought remarkable changes on crystallization and relaxation behaviors. So a comprehensive investigation on the structure–property relationship is necessary.

Fig. 7 gives the notched Charpy impact strength of annealed IPCs with different β-NA contents. It is noted that for the unannealed samples, their impact strength is effectively improved by adding β-NA at room temperature but it is enhanced slightly at 0 °C. However, impact properties of all IPC samples subjected to annealing treatment deteriorate. Generally, β crystal modification is an effective way to toughen PP because the energy could be much dissipated by deformation of looser crystallographic texture.¹⁵ The results of crystallization behavior above suggested that the content of β crystals increases with adding more β-NA but decreases after annealing. Thus, the toughness is changed with the variation of β crystal content. The structure and behaviors in the amorphous phase also play important roles in toughness as the energy could be dissipated by the relaxation behaviors of molecular chains. The fracture resistance is the substantial response of chain mobility. However, the effect of slightly increased chain mobility during annealing which may enhance the toughness would be compensated by the reduction of β crystal. So the impact strength tested at room tends to decrease after annealing. Nevertheless, the toughness at low temperature of 0 °C is almost unchanged by adding β-NA or undergoing annealing treatment. Even its dependence on β crystal content at 0 °C is invalid. It indicates that the presence of β crystal has little effect at low temperature (below T_g) for IPC samples. Few chains in HPP amorphous phase is activated at this temperature so that the impact energy can't be transmitted effectively by crystal deformation.

Fig. 7 Notched Charpy impact strength of neat IPC and β-nucleated IPC (0.1 wt%) tested at 0 °C and 23 °C.

It is widely accepted that the major toughening mechanisms of elastomer-toughened polymer systems are: multiple crazing, shear-yielding theory, microvoids and cavitation theories.^10,44,45 Here, impact strengths of IPC are strongly dependent on features of soft dispersed particles (size and content) and matrix. The dispersed particles can generate crazes or undergo internal cavitation due to stress concentration, triggering the shear yielding and plastic deformation of the HPP matrix around them as well as ending the propagation of cracks. However, the occurrence of shear yielding or plastic deformation of crystallizable HPP matrix is more efficient for the consumption of energy. Given that the extent of shear yielding is related to the chain mobility in HPP amorphous phase, the loose β crystal is favorable for the improvement of fracture toughness compared with α crystal. In this work, the contribution of dispersed particles on toughness in IPCs is similar since the content, size and chain mobility of them (see Fig. 4 and 5) are close. The difference of toughening effect may mainly come from toughness of HPP matrix, which is treated with different contents of β-NA or undergoing different annealing conditions. This deduction accords well with inserted picture in Fig. 7. The intensive and extensive stress whitening in neat IPC, which results from the decrease in refraction index of crazes, implies the occurrence of crazing. As to β-nucleated IPCs, they consume energy with a more efficient way, by shear yielding and plastic deformation. So a smaller whitening zone can be observed.

To explore the toughening mechanism deeply, the micrographs of impact fracture section were shown in Fig. 8. For neat IPC, smooth surface and deep crazes can be viewed at fracture section, which could be ascribed to brittle damage by crazing. However, plastic deformation bands and cavitation are distinct in β-nucleated IPC and they weaken with rising annealing temperature. It means that looser β crystal should be beneficial for not only relieving stress by sliding deformation, but also delivering breaking stress to form cavitation. The voids in β-nucleated IPC could help to ease craze formation and multiply shear yielding. So, obvious plastic deformation areas can be observed in β-nucleated IPC while extensive stress whitening phenomenon appears in neat IPC. But annealing can induce the β–α transition of crystals. As a result, the impact property gets worse with the decreasing content of β crystals.

Fig. 8 SEM micrographs of impact fracture section for neat IPC (a) and β-nucleated IPC (0.1 wt%) (b)–(d) with different annealing treatment. (b) Unannealed; (c) 124 °C, 30 min; (d) 140 °C, 30 min.

However, our results are contrary to some work reported. Fu et al.^11,12,46 believed that annealing treatment would promote chain mobility in β-nucleated IPC so that the toughness was improved effectively and the toughness enhancement only occurs at testing temperature of 0 °C rather than at room temperature. Here, the good toughening effect is obtained by introduction of β-NA at room temperature but the toughness becomes worse sharply after annealing. As mentioned above, the toughness of IPC is related to dispersed particles (size and content) and matrix. Wu et al.³⁶ reported that the optimum diameter of dispersed phase for toughening was close to the critical size, where a sharp brittle/tough transition occurred at a constant amount of dispersed phase. Large particles are more effective in initiating the crazing while small ones are more effective in initiating the yielding. By contrast, yielding is a better way than crazing in dissipating energy. The large particles caused poor original toughness. In Fu's work,^11,12,46 the sizes of particles are 0.5–0.8 μm (the critical size is about 0.5 μm for rubber-modified PP materials⁴⁷) and the great original impact strength may cover the effects of other factors. But in our work, the average sizes of dispersed particles are about 1.0–1.5 μm, which is much greater than critical size. The effect of crystallization induced by annealing treatment on toughness may be expressed significantly.

To verify our explanation, we prepared β-nucleated IPC samples with smaller dispersed particles. The average sizes of particles are about 0.7–0.9 μm. The impact strength tested at room temperature of 23 °C and 0 °C are shown in the Fig. 9 and both of them are greater than that in Fig. 7 under the same condition. At this moment, the toughness just reduces slightly with increasing annealing temperature not until that it goes down sharply at high temperature of 150 °C. The smaller particles are more effective in toughening so that the changes of crystallization behavior caused by annealing here play little effect on the reduction of the toughness. This result is just in accordance with the above speculation.

Fig. 9 Notched Charpy impact strength of IPC with β-NA content of 0.1 wt%.

4. Conclusions

With increasing the content of β-NA in IPC, the total crystallinities of α and β phase rise slightly but the proportion of β crystal increases greatly. Meanwhile, comparing with neat IPC, the chain mobility in RAF and MAF were both strengthened markedly by adding β-NA. That may be ascribed to the loose structure of β crystals which reduced the restriction of chain segments. When IPC samples were annealed between T_g and T_m, thickening and perfection of lamellae occurred, resulting in a β–α transition of crystals. The chain mobility in RAF was constrained further by lamellar thickening while that in MAF was slightly enhanced due to the decrease of MAF density. The excellent toughness in β-nucleated IPC was mainly attributed to shear yielding and deformation of loose β crystals as well as cavitation that consumed most impact energy. By contrast, the stable α crystals in neat IPC couldn't deform and slip easily so that the impact energy was mostly absorbed by crazing induced by dispersed particles. As a result, distinct stress whitening and crazes were observed in impact specimens of neat IPC. The results suggested that the toughness became worse by annealing treatment. It is thought that the decreasing β crystal of HPP was responsible for deterioration.

Acknowledgements

This work was supported by National Nature Science Foundation of China (No. 51573163, 51173157), Zhejiang Provincial Natural Science Foundation of China (No. R16E030003) and the Hujiang Foundation of China (No. B14006).

References

1. C. H. Zhang, Y. G. Shangguan, R. F. Chen and Q. Zheng, J. Appl. Polym. Sci., 2011, 119, 1560–1566.
2. C. H. Zhang, Y. G. Shangguan, R. F. Chen, Y. Wu, F. Chen, Q. Zheng and G. H. Hu, Polymer, 2010, 51, 4969–4977.
3. S. J. Song, J. C. Feng, P. Y. Wu and Y. L. Yang, Macromolecules, 2009, 42, 7067–7078.
4. H. S. Tan, L. Li, Z. N. Chen, Y. H. Song and Q. Zheng, Polymer, 2005, 46, 3522–3527.
5. B. W. Qiu, F. Chen, Y. Lin, Y. G. Shangguan and Q. Zheng, Polymer, 2014, 55, 6176–6185.
6. Z. Q. Fan, Y. Q. Zhang, J. T. Xu, H. T. Wang and L. X. Feng, Polymer, 2001, 42, 5559–5566.
7. C. Tong, Y. Lan, Y. Chen, Y. Chen, D. Yang and X. Yang, J. Appl. Polym. Sci., 2012, 123, 1302–1309.
8. F. Chen, Y. Shangguan, Y. Jiang, B. Qiu, G. Luo and Q. Zheng, Polymer, 2015, 65, 81–92.
9. F. Chen, B. W. Qiu, Y. G. Shangguan, Y. H. Song and Q. Zheng, Mater. Des., 2015, 69, 56–63.
10. J. Z. Liang and R. K. Y. Li, J. Appl. Polym. Sci., 2000, 77, 409–417.
11. F. Luo, C. Xu, K. Wang, H. Deng, F. Chen and Q. Fu, Polymer, 2012, 53, 1783–1790.
12. H. Bai, F. Luo, T. Zhou, H. Deng, K. Wang and Q. Fu, Polymer, 2011, 52, 2351–2360.
13. L. F. Ma, X. F. Wei, Q. Zhang, W. K. Wang, L. Gu, W. Yang, B. H. Xie and M. B. Yang, Mater. Des., 2012, 33, 104–110.
14. J. Varga, J. Mater. Sci., 1992, 27, 2557–2579.
15. J. Varga, J. Macromol. Sci., Part B: Phys., 2002, 41, 1121–1171.
16. C. Grein, C. J. G. Plummer, H. H. Kausch, Y. Germain and P. Béguelin, Polymer, 2002, 43, 3279–3293.
17. H. Bai, Y. Wang, Z. Zhang, L. Han, Y. Li, L. Liu, Z. Zhou and Y. Men, Macromolecules, 2009, 42, 6647–6655.
18. C. Grein, K. Bernreitner and M. Gahleitner, J. Appl. Polym. Sci., 2004, 93, 1854–1867.
19. P. Maiti, M. Hikosaka, K. Yamada, A. Toda and F. Gu, Macromolecules, 2000, 33, 9069–9075.
20. C. Hedesiu, D. E. Demco, R. Kleppinger, G. V. Poel, W. Gijsbers, B. Blümich, K. Remerie and V. M. Litvinov, Macromolecules, 2007, 40, 3977–3989.
21. X. Zhou, J. Feng, J. Yi and L. Wang, Mater. Des., 2013, 49, 502–510.
22. F. Ramsteiner, G. E. McKee and M. Breulmann, Polymer, 2002, 43, 5995–6003.
23. W. L. Zhao, Chinese Pat., CN 102181092A, 2011.
24. J. X. Li and W. L. Cheung, J. Mater. Process. Technol., 1997, 63, 472–475.
25. Q. Zheng, Y. G. Shangguan, S. K. Yan, Y. H. Song, M. Peng and Q. B. Zhang, Polymer, 2005, 46, 3163–3174.
26. H. Huo, S. Jiang, L. An and J. Feng, Macromolecules, 2004, 37, 2478–2483.
27. R. H. Somani, B. S. Hsiao, A. Nogales, H. Fruitwala, S. Srinivas and A. H. Tsou, Macromolecules, 2001, 34, 5902–5909.
28. Y. G. Shangguan, Y. H. Song, M. Peng, B. Li and Q. Zheng, Eur. Polym. J., 2005, 41, 1766–1771.
29. J. Karger-Kocsis and P. P. Shang, J. Therm. Anal. Calorim., 1998, 51, 237–244.
30. G. Y. Shi, B. Huang and J. Y. Zhang, Macromol. Rapid Commun., 1984, 5, 573–578.
31. Y. M. Liu, Z. Z. Tong, J. T. Xu, Z. S. Fu and Z. Q. Fan, J. Appl. Polym. Sci., 2014, 131, 40753–40763.
32. T. S. Motsoeneng, A. J. van Reenen and A. S. Luyt, Polym. Int., 2015, 64, 222–228.
33. T. Y. Cho, B. Heck and G. Strobl, Colloid Polym. Sci., 2004, 282, 825–832.
34. G. W. H. Höhne, Polymer, 2002, 43, 4689–4698.
35. S. Wu, J. Appl. Polym. Sci., 1988, 35, 549–561.
36. S. Wu, Polym. Eng. Sci., 1990, 30, 753–761.
37. M. Hoyos, P. Tiemblo and J. M. Gómez-Elvira, Polymer, 2007, 48, 183–194.
38. Y. Men and G. Strobl, Polymer, 2002, 43, 2761–2768.
39. B. Qiu, F. Chen, Y. Shangguan, L. Zhang, Y. Lin and Q. Zheng, RSC Adv., 2014, 4, 58999–59008.
40. J. K. Kim, D. J. Park, M. S. Lee and K. J. Ihn, Polymer, 2001, 42, 7429–7441.
41. F. Auriemma, C. De Rosa, R. Di Girolamo, A. Silvestre, A. M. Anderson-Wile and G. W. Coates, J. Phys. Chem. B, 2013, 117, 10320–10333.
42. G. Machado, E. J. Kinast, J. D. Scholten, A. Thompson, T. D. Vargas, S. R. Teixeira and D. Samios, Eur. Polym. J., 2009, 45, 700–713.
43. K. Yamada, M. Hikosaka, A. Toda, S. Yamazaki and K. Tagashira, Macromolecules, 2003, 36, 4790–4801.
44. Y. Lin, H. Chen, C. M. Chan and J. Wu, Macromolecules, 2008, 41, 9204–9213.
45. Y. Li, G. X. Wei and H. J. Sue, J. Mater. Sci., 2002, 37, 2447–2459.
46. F. Luo, C. Xu, N. Ning, K. Wang, H. Deng, F. Chen and Q. Fu, Polym. Int., 2013, 62, 172–178.
47. B. Z. Jang, D. R. Uhlmann and J. B. V. Sande, J. Appl. Polym. Sci., 1985, 30, 2485–2504.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra01046f

---