Suppressing of γ-Crystal Formation in Metallocene-Based Isotactic

Apr 15, 2012 - ACS eBooks; C&EN Global Enterprise .... *(Z.-M.L.) Tel/Fax: +86-28-8540-6866. ... Fax: +1-631-632-6518. .... Published online 15 April ...
0 downloads 0 Views 3MB Size
Download PDF
Article pubs.acs.org/JPCB

Suppressing of γ-Crystal Formation in Metallocene-Based Isotactic Polypropylene during Isothermal Crystallization under Shear Flow Yan Wang,† Chen Chen,‡ Jia-Zhuang Xu,† Jun Lei,† Yimin Mao,§ Zhong-Ming Li,*,† and Benjamin S. Hsiao*,§ †

College of Polymer Science and Engineering and State Key Laboratory of Polymer Materials Engineering, Sichuan University, Chengdu 610065, P. R. China ‡ Analytical and Testing Center, Sichuan University, Chengdu 610065, P. R. China § Department of Chemistry, Stony Brook University, Stony Brook, New York 11794-3400, United States S Supporting Information *

ABSTRACT: The effect of shear flow on isothermal crystallization behavior of γ-crystals in metallocene-based isotactic polypropylene melt was investigated by in situ synchrotron wide-angle X-ray diffraction (WAXD). In the sample under weak shear (at strain of 300% for 30 s duration), simultaneous evolution of α- and γ-crystals occurred, and the final fraction of γ-crystals (fγ) was 0.66, which was identical to the undeformed sample (PP-Static). In this scenario, α-crystals probably served as effective seeds for nucleation of γ-crystals. In the samples under strong shear (at strain of 500% for 30 s duration or long-time continuous shear at strains of 100% and 500%), the sequential emergence of α- and γ-crystals was observed. In this case, molten polymer chains were probably constrained by the surrounding crystals after intense short-time shear and/or maintained their extended chain conformation after long-time shear. These oriented chains had little chance to form the γ-crystals directly, behaving very differently from the relaxed chains. Under strong shear fields, the emergence of γ-crystals was delayed or inhibited, whereas the fγ value was also decreased rapidly. A simple model for the possible pathway of γ-crystal formation in the strong shear environment was proposed.

■

the relationship between the flow field and polymorphism in m-iPP is still elusive up to date. To be specific, the coexistence of α- and γ-crystals in m-iPP has been reported, but the effect of flow on the formation of γ-crystals is still unclear. Thus far, the subject related to the formation of γ-crystals has been mainly focused on the quiescent crystallization of m-iPP.3,4,7,16,23,24 Generally, the crystallization of γ-crystals is thermodynamically favored in m-iPP, especially when samples contain an appreciable amount of stereodefects. The content of γ-crystals can be increased at higher isothermal crystallization temperatures or lower cooling rates under nonisothermal crystallization conditions.7,16,24 It is also known that the γ-crystal structure is closely associated with α-crystals through epitaxial crystallographic association.25−27 It is interesting to note that several groups have reported very different findings about the effect of flow on the formation of γ-crystals. For example, Hsiao et al. investigated the correlation between the shear flow and/or oriented α-crystals and the content of γ-crystals formed in long-chain branched iPP (LCBiPP).28 Their results indicated that the higher content of γ-crystals in high LCB-level samples under shear could be

INTRODUCTION Isotactic polypropylene (iPP) has a simple molecular architecture, but it possesses several polymorphic structures upon crystallization.1−8 The polymorphism in iPP is closely related to its mechanical performance and thus has attracted a great deal of scientific interests.9−14 One major interest in this material is how the polymorphism can be influenced by chain microstructure in combination with applied flow fields.7,8,15 Traditional Ziegler−Natta based iPP (ZN-iPP) generally crystallizes into α-crystals, while metallocene-based iPP (m-iPP) has a tendency to crystallize into γ-crystals.3,7,16,17 The discrepancy between ZN-iPP and m-iPP samples can be attributed to the different type and distribution of insertion defects along the chains. In ZN-iPP, most defects are segregated into small fractions of chain segments that are difficult to crystallize. This would result in long isotactic sequences and favor the crystallization of α-crystals. In contrast, the distribution of defects in m-iPP is relatively random, which would significantly reduce the isotactic sequence length and result in higher content of γ-crystals. The effect of flow field on the polymorphism of ZN-iPP (e.g., α- and β-crystals) has been well documented.1,18−21 It is known that the flow (shear or elongation) first induces oriented α-crystal nuclei, while the surface of oriented α-crystals can subsequently trigger the formation of β-crystals.1,21,22 However, © 2012 American Chemical Society

Received: January 10, 2012 Revised: March 5, 2012 Published: April 15, 2012 5056

dx.doi.org/10.1021/jp3003068 | J. Phys. Chem. B 2012, 116, 5056−5063

The Journal of Physical Chemistry B

Article

mainly attributed to the higher region-defects in the chains and seemed to be independent of the shear-induced molecular orientation. In contrast, Hong and Seo studied a terpolymer of poly(propylene-co-ethylene-co-1-butene), consisting of 94.5 wt % of polypropylene unit, under a mixed elongational and shear flow during the foaming process.8 They concluded that the γ-crystal content tended to increase with the foam expansion ratio (larger expansion ratio means stronger elongational flow). Similarly, Kalay et al. pointed out that the high fraction of γ-crystals might be due to the pronounced molecular orientation in the injection molded ZN-iPP.29 Recently, our group also determined the spatial distribution of the γ-crystals in the injection molded m-iPP samples by performing the oscillatory shear injection molding experiment,30 which involved the solidification of polymer melt under a continuous and oscillatory shear. Our results indicated that γ-crystals were somewhat depressed by the continuous shear, which is not reconciled with any of the findings mentioned above.8,28,29 It is clear that in real-world polymer processing conditions, the complicated flow fields accompanied by nonisothermal crystallization environments make it difficult to decouple the factors influencing for the formation of γ-crystals. Therefore, there is a need to carry out experiments that can uncover the underlying relationship between flow and formation of γ-crystals; in particular to address the question “How does a flow field influence the evolution of γ-crystals in m-iPP?” This question is of high importance because the flow-induced crystalline structure change can strongly affect the final performances of the shaped iPP part.9,15 To address the above question, the in situ isothermal crystallization study of m-iPP was carried out at 140 °C under shear flow. The temperature of 140 °C was selected because it is in the vicinity of the nominal melting point of m-iPP (Tm = 145 °C). This experimental temperature allowed us to observe flow-induced crystallization at the very early stage.14,24 Two different shear protocols were carried out: a short-time shear flow and a long-time continuous shear flow (both in the oscillatory mode). The long-time continuous shear means that the shear flow continued until polymer melt was solidified. This protocol is useful to mimic the real-world processing conditions, such as the oscillatory shear injection molding.30,31 In situ synchrotron wide-angle X-ray diffraction (WAXD) was used to monitor the structure changes during crystallization. On the basis of the evolution of crystal structure, the influence of shear flow on the formation of γ-crystals was revealed. A possible evolution mechanism of γ-crystals formation under the oscillatory shear samples was thus proposed accordingly.

Table 1. Metallocene-Based Isotactic Polypropylene Sample Information sample

Mw (g/mol)a

Mw/ Mna

Tm (°C)b

[mr] %c

[rr] %c

[mm] %c

562N

407,000

2.25

145

3.09

0.386

96.05

a

Mw and Mw/Mn of m-iPP was obtained by gel permeation chromatography (GPC) in 1,2,4-trichlorobenzene at 160 °C. b Determined from the peak position of the DSC curves recorded at a heating rate of 10 °C/min. cThe content of triad stereosequences are obtained from the 13C NMR spectra. The was sample recorded on a Bruker Avance 400 operating at 500 MHz in the Fourier transform mode of 10% w/v polymer solutions in deuterated 1,2,4-trichlorobenzene and benzene at 120 °C.

A Linkam CSS-450 high-temperature shear stage modified for an in situ X-ray scattering study was used to control the shear flow and thermal history of the samples. The polymer samples were in the form of a ring (the outer diameter = 20 mm and the inner diameter = 10 mm) cutting from the compression-molded sheet (about 0.5 mm thick) and were mounted between two X-ray windows (diamond and Kapton film) in the Linkam shear stage. The gap between the two windows was adjusted to the sample thickness. The chosen temperature protocol was as follows: (a) heat the polymer sample from room temperature to 200 °C at a rate of 30 °C/ min; (b) keep the polymer melt by holding the temperature at 200 °C for 5 min; (c) cool down to 140 °C at a rate of 30 °C/ min; (d) WAXD data collection was carried out right after the sample reaches 140 °C. The entire crystallization process was monitored in real time. To apply the controlled shear, polymer melts were subjected to varying flow conditions after the temperature reached 140 °C as shown in Figure 1: (a) short-time shear mode, shear strain of 300% and 500% at the oscillation frequency of 2 Hz for 30 s, designated as PP-300-30, PP-500-30, respectively; (b) long-time continuous shear mode, shear strain of 100% and 500% at the oscillation frequency of 2 Hz in which the shear continued until the melt was solidified, designated as PP-100-L, PP-500-L, respectively. For comparison, quiescent crystallization at 140 °C after cooling from 200 to 140 °C, designated as PP-Static, was also investigated. WAXD Data Analysis. 2D-WAXD patterns are integrated over the entire azimuthal angle to convert to 1D diffraction profiles (the missing diffraction intensity from the flat plate geometry and the orientation effect were corrected accordingly). This 1D profile illustrated the integrated intensity as a function of |q| = 4π sin θ/λ, where q is the absolute value of the scattering vector, λ being the wavelength of X-ray beam (1.371 Å), and 2θ is the scattering angle. The Gaussian function was used to fit each peak in the diffraction profile, where the fitting method allowed us to separate the contributions of crystalline and amorphous fractions.1 The relative crystallinity index (Xc) was calculated by

■

EXPERIMENTAL SECTION Materials. Isotactic polypropylene (iPP), polymerized by metallocene catalyst, was supplied by the Basel Company. The sample information, such as molecular weight (Mw), polydispersity (Mw/Mn), nominal melting temperature (Tm), and content of triad stereosequences, is shown in Table 1.30 In Situ Wide-Angle X-ray Diffraction Measurements. WAXD measurements were carried out at the Advanced Polymer Beamline (X27C) in the National Synchrotron Light Source (NSLS), Brookhaven National Laboratory (BNL). A 2D MAR CCD X-ray detector (MARUSA) was employed for the detection of 2D-WAXD images, with an image resolution of 512 × 512 pixels (pixel size = 158 μm). The sample-to-detector distance was 75 mm. The data acquisition time for each scattering pattern (image) was 30 s.

Xc =

∑ Acryst ∑ Acryst + ∑ A amorp

(1)

where Acryst and Aamorp represent the fitted areas contributed by crystals and amorphous phase, respectively. The normalized crystallinity Xr(t) was used to estimate crystallization kinetics

X r (t ) = 5057

X c (t ) X c (∞ )

(2)

dx.doi.org/10.1021/jp3003068 | J. Phys. Chem. B 2012, 116, 5056−5063

The Journal of Physical Chemistry B

Article

(111)α, and (041)α for α-crystals;33 and to (111)γ, (008)γ, (117)γ, (202)γ, and (026)γ for γ-crystals.34 The difference in the diffraction curves of α- and γ-crystals is the (130)α peak that can only be observed for the α-crystals, and the (117)γ peak that can only be observed for the γ-crystals (the existence of γ-crystals was marked by the vertical solid line in Figure 2). Although α- and γ-crystals appeared in all the samples, the induction times and the contents for these two crystals were very different. An induction period for crystallization was observed for all the samples, where the induction times for α- and γ-crystals were not always the same. The induction time for α- and γ-crystals was denoted as tiα and tiγ, respectively. In this study, the PP-Static sample (Figure 2a) showed a very long induction time for crystallization, whose diffraction peaks first appeared at tiα = tiγ = 1950 s with a very low intensity. A sharp decrease of induction time (from 1950 to 270 s) was observed after the application of weak shear with duration of 30 s and strain of 300% (PP-300-30) (Figure 2b). During isothermal crystallization of m-iPP, the simultaneous occurrence of α- and γ-crystals has been reported before.16,17,35 The same behavior was also observed in the sheared m-iPP sample under weak shear conditions (PP-300-30). As the shearing intensity increased (e.g., the strain increased to 500% for 30 s (PP-50030), the crystalline signals appeared immediately upon cessation of shear (t = 30 s, as seen in Figure 2c). It is very interesting to note that the α-crystals developed first at tiα = 30 s, and the characteristic diffraction peaks of γ-crystals became visible subsequently at tiγ = 120 s (Table 2). This suggests that a critical value of shear intensity (between 300% and 500%) for the short-time shear mode might exist, under which the transition took place from simultaneous formation to sequential formation of α- and γ-crystals. To our knowledge, this is the first time that one observed the separate emergence of α- and γ-crystals at the beginning of crystallization. It seems that the evolution of a different crystalline structure varies with the shear intensity. In order to reveal the impact of strong shear on the formation of crystalline structure, the long-time continuous shear flow mode was further imposed on the m-iPP melt. It is seen that, when the long-time continuous shear flow at a strain of 100% and 500% (PP-100-L and PP-500-L) (Figure 2d,e) was applied, the induction time for α-crystals (tiα) was 450 and 30 s, respectively, while the presence of γ-crystals (tiγ) lagged significantly behind, with an induction time of 900 and 1050 s, respectively (Table 2). These results indicate that the long-time continuous shear flow enhances the formation of α-crystals but inhibits the formation of γ-crystals at the early stage of m-iPP crystallization. After the induction period, the diffraction intensities of both α- and γ-crystals in all the samples increased with time until the completion of crystallization. Using eqs 1 and 2, the normalized crystallinity was calculated and the corresponding time evolution trend was shown in Figure 3. Compared to the quiescent experiment, the crystallization was accelerated by the application of shear in two fashions: (1) the parallel translation of the kinetics curve, i.e., the crystallization curve shifts due to the reduced induction time (e.g., the static and weakly sheared m-iPP samples (PP-Static and PP-300-30)) where all curves exhibit the same sigmoidal shape; (2) the shape change of the kinetics curve, i.e., not only the crystallization curve shifts to a shorter time, its shape also becomes different (e.g., as seen in the strongly sheared m-iPP samples, PP-500-30, PP-100-L, and PP-500-L). These kinetic curves were analyzed using the

Figure 1. Schematics of temperature and shear protocol: (a) shorttime shear mode, (b) long-time continuous shear mode.

where Xc(t) and Xc(∞) represent the crystallinity at a given time t during crystallization and that at the final stage of crystallization, respectively. The fraction of the γ-crystals was estimated using eq 3, as suggested by Turner-Jones et al.32 fγ = Iγ(117)/[Iγ(117) + Iα(130)]

(3)

where Iγ(117) and Iα(130) represent the areas of the corresponding diffraction peaks. The crystallinity of γ-crystals (Xγ) and the crystallinity of α-crystals (Xα) are given by X γ = f γXc

(4)

Xα = 1 − X γ

(5)

For convenience, normalized crystallinity of γ- and α-crystals is designated as Xrγ and Xrα, respectively.

■

RESULTS Selected 2D-WAXD patterns of m-iPP samples under varying shear and quiescent states during isothermal crystallization at 140 °C are illustrated in the Supporting Information. These patterns give the reference time of the emergence of different crystal structures under varying experimental conditions. The time evolution of the corresponding 1D-WAXD curves was shown in Figure 2. Taking Figure 2a, for example, the observed diffraction peaks can be assigned to (110)α, (040)α, (130)α, 5058

dx.doi.org/10.1021/jp3003068 | J. Phys. Chem. B 2012, 116, 5056−5063

The Journal of Physical Chemistry B

Article

Figure 2. 1D-WAXD intensity profiles versus scattering vector q of PP-Static (a), PP-300-30 (b), PP-500-30 (c), PP-100-L (d), and PP-500-L (e) isothermally crystallized at 140 °C. The red lines refer to (1) image of amorphous melt, (2) image showing emergence of the α-crystals, (3) image showing emergence of γ-crystals reflections, and (4) images showing the final diffraction profile after isothermal crystallization.

about 1.45 for the shape changing type. This suggests that the crystal growth for (1) and (2) types can be attributed to the 3D-growth and the growth of fibrillar and/or disk-like crystals, respectively. These results indicate that shear-induced molecular orientation promotes the development of oriented crystalline structure. Figure 4 illustrates the fraction of γ-crystals (fγ) as a function of crystallization time. It is seen that fγ of PP-Static was about 0.66 at the beginning of crystallization and stayed constant through out the entire crystallization process. The initial fγ value of PP-300-30 was about 0.58 (from t = 270 to 1000 s) and then reached a plateau value at about 0.66 (from t = 1000 to 2700 s) that was exactly the same as that of the PP-Static sample. This result confirms that both α- and γ-crystals display the identical increase in the induction time. However, for all PP-500-30,

Table 2. Characteristic Time for the Structure Development under Different Crystallization Conditionsa PP-Static PP-300-30 PP-500-30 PP-100-L PP-500-L

tiα (s)

tiγ (s)

tpα (s)

tpγ (s)

1950 270 30 450 30

1950 270 120 900 1050

5580 2250 600 870 870

5580 2250 1300 1200 1200

a tiα, induction time of α-crystals; tiγ, induction time of γ-crystals; tpγ, emergence time of the plateau of γ-crystals.

Avrami plot to investigate the nature of the crystal growth. The results indicate that the Avrami exponent (n) was 3.62 for the kinetics curves having the parallel translation fashion and was 5059

dx.doi.org/10.1021/jp3003068 | J. Phys. Chem. B 2012, 116, 5056−5063

The Journal of Physical Chemistry B

Article

Figure 3. Development of normalized crystallinity at 140 °C.

Figure 5. Crystallinity evolution of γ-crystals (a) and α-crystals (b) at 140 °C.

Figure 4. Development of the fraction of γ-crystals ( fγ) at 140 °C.

important role in the formation of γ-crystals. The underlying mechanism for this phenomenon will be discussed next.

PP-100-L, and PP-500-L samples, the fγ value increased very gradually with time and could only reach a low plateau value. This indicates that the formation of γ-crystals is not preferred under the strong shear conditions, whereas the α-crystals become favored at the expense of γ-crystal content. The comparative evolution of γ- and α-crystal crystallinity was illustrated in Figure 5. It is found that the development of α-crystals exhibited the same trend as that of γ-crystals for PP-Static and PP-300-30, whereas the development of α-crystals became the dominant event over the development of γ-crystals for PP-500-30, PP-100-L, and PP-500-L samples. From the crystal evolution profiles, the saturation for two kinds of crystals (i.e., tpα and tpγ, the times to reach the plateau crystallinity) were estimated. The results are listed in Table 2. For PP-Static and PP-300-30 samples, tpα and tpγ exhibited the same value. However, for PP-500-30, PP-100-L, and PP-500-L samples, tpγ became longer than tpα (Table 2). Furthermore, it is interesting to note that the appearance of γ-crystals (tiγ) for PP-100-L and PP-500-L samples, which occurred at 900 and 1050 s, respectively, was longer than tpα (870 s). This suggested that only when α-crystals become saturated, γ-crystals would appear in the long-time continuous shear mode samples. Compared to PP-Static, the fraction of γ-crystals in PP-500-30 was found to depress from 66% to 50% and was further reduced to 7% when long-time continuous shear was applied. Considering the above results, one can conclude that the shear flow regulates the polymorphism, specifically, the manner of forming α- and γ-crystals and their corresponding contributions in m-iPP. The melt conditions adjusted by shear-induced chain orientation and associated relaxation behavior play an

■

DISCUSSION On the basis of the above results, we provided the first evidence that the application of shear could delay or even inhibit the formation of γ-crystals in m-iPP under isothermal conditions. The general behavior from our study suggests that, under the short-time shear (PP-300-30), i.e., below the critical shear strength, shear has no effect on the content of γ-crystals ( fγ) compared with the static sample. In contrast, the higher shear strength (PP-500-30) can result in the emergence of γ-crystals at a later time than that of α-crystals, as well as the reduced value of fγ. The long-time continuous shear (PP-100-L and PP500-L) can further delay the formation of γ-crystals and depress the fγ value severely. These results are closely related with the crystallization mechanism of γ-crystals in the sheared melt. In the static sample, the simultaneous emergence of α- and γ-crystals can be associated with the likelihood of α-crystals acting as effective seeds to nucleate and grow γ-crystals.25−27 In other words, once the α-crystal is generated, an epitaxial growth of γ-crystal begins immediately on the surface of the α-crystal. In this case, simultaneous evolution of α- and γ-crystals occurs, and the two crystal forms reach the saturation point simultaneously. This explanation is certainly applicable to the PP-300-30 sample, which is a characteristic of accelerating the crystallization process without changing the fγ value (Figures 4 and 5a). This is also consistent with the hypothesis that, under the relative weak shear deformation (i.e., the Weissenberg Number, We < 1 from Supporting Information), the density of stable nuclei can increase, but the 5060

dx.doi.org/10.1021/jp3003068 | J. Phys. Chem. B 2012, 116, 5056−5063

The Journal of Physical Chemistry B

Article

surrounding preoriented molecular chains would relax back to the original coil state. As a result, the content of γ-crystal ( fγ) under weak shear should be the same as that of the static melt, which was seen in this study. However, this is not the case for m-iPP under the stronger shear (PP-500-30) or long-time continuous shear (PP-100-L and PP-500-L). For PP-500-30 (i.e., We > 1 from Supporting Information), the α-crystal content was found to be around 5% (Figure 5b) at the end of shear duration (30 s), which is consistent with the study of Balzano who induced crystals during a brief shear but with a high degree of orientation.36 It is conceivable that there are a certain fraction of chains immobilized by their attachment to the adjacent crystalline structure, whereby these chains cannot be relaxed freely, and would remain oriented.37 Such segmental ordering would increase the likelihood that they can be arranged into α-crystal structure with the parallel chain configuration. Thus, the formation of α-crystals can prevail in the early stages of crystallization due to these oriented chains. However, the residual chains (less than 30 s relaxation time, see Supporting Information) that are not constrained by the surrounding crystals can relax back to their equilibrium coil state. As a result, both α-crystals with the parallel chain configuration and γ-crystals with the nonparallel chain configuration can be formed simultaneously. It is possible that the plateau value of fγ for PP-500-30 (about 0.50, which is slightly lower than that of PP-Static (fγ = 0.66)), is the result of the preference for formation of α-crystals in the very early stage of crystallization. When the m-iPP melt sample was subjected to long-time continuous shear (PP-100-L and PP-500-L), the steady-state hydrodynamic drag forces across the molecules can overcome the entropic barrier and stretch the chains along the shear axis. As in the PP-500-30 sample, the stretched chains are more likely to rearrange themselves into α-crystals at the early stages of crystallization. The long-time continuous shear thus can enable the longer chains to stretch into the extended state. In this case, the formation of α-crystals become preferred, whereas the formation of γ-crystals only takes place when α-crystals reach the plateau value. It is reasonable to conclude that a fraction of the long chain can stay oriented along the flow direction, where the rest of the chains (as those not attached to the crystals) can relax. Only the relaxed chains have a chance to form the γ-crystals with the nonparallel chain configuration. In addition to the above argument, there is an alternative explanation. That is at the late stages of crystallization, the sample skin becomes solid where the stress would not be able to penetrate into the internal part of the molten samples. In this case, the chains that are not involved in the crystal regions can relax. It is possible that both scenarios may occur. Finally, we argue that only oriented chains can be involved in the formation of α-crystals, where the relaxed chains after the saturation of α-crystals would lead to the formation of γ-crystals. This may be the reason why the long-time continuous shearing gave rise to a greater reduction of the γ-crystals content (Xrγ = 0.25 and 0.07 for PP-100-L and PP-500-L, respectively) than that of PP-500-30 (Xrγ = 0.5). A simple model can be used to describe the process of shearsuppressed formation of γ-crystals in the m-iPP melt, as schematically illustrated in Figure 6. To rationalize this model, it is necessary to briefly review the lattice difference between the α- or γ-crystals. The α-crystal has a monoclinic unit cell, with polymer chains aligned parallel to the c axis.33 The unit cell of the γ-crystals is orthorhombic, where its c axis is particularly

Figure 6. Schematic diagrams of entangled m-iPP chains under deformation. (a) The flexible polymer chains in the entangled state without deformation. (b) The entangled polymer chains under deformation. (c) The oriented chains favor the formation of α-crystals (indicated as the red lamellae). (d) The relaxed chains can be rearranged into both α- and γ-crystals (as indicated by the blue lamellae).

large due to the fact that chains are piled along this direction in a bilayered fashion (i.e., each layer contains two chains). There is an angle of 80° between the two adjacent layers, indicating that each layer occupies one of the diagonals of the ab plane of the unit cell with a cross-hatching style.34,38 In the random coil state, as in the starting melt (Figure 6a), the flexible polymer chains are entangled. Under shear flow, the entangled chains would deform in a cooperation manner when the hydrodynamic friction across the chain exceeds the entropic elasticity of the coiled chains (Figure 6b). Different scenarios can occur after this event. First, some oriented chains can survive, and the crystal growth would take place by rearranging the oriented chains into only α-crystals with the parallel chain configuration (as in Figure 6c). However, some stretched chains (e.g., the chains not constrained by the adjacent crystals) can relax back to the initial coiled state. These relaxed chains can be reconstructed into both α-crystals with the parallel chain configuration and γ-crystals with the nonparallel chain configuration (as shown in Figure 6d). Depending on the molecular attributes of m-iPP, the ratio of α- and γ-crystals should be similar to that in the unsheared melt. In this model, the pathway for formation of the γ-crystals in the sheared melt is slightly different from the traditional mechanism, i.e., α-crystals act as effective seeds to nucleate γ-crystals, where they grow simultaneously. Our results indicate they may grow subsequently under the strong shear conditions. We note that the highly oriented chains have little chance to form γ-crystals, while the relaxed chains in the coil state will be the major source to form γ-crystals. The present study clearly 5061

dx.doi.org/10.1021/jp3003068 | J. Phys. Chem. B 2012, 116, 5056−5063

The Journal of Physical Chemistry B

Article

reveals the inhibiting affect of shear flow on the formation of γ-crystals at the molecular level. In other words, we argue that the concept about the preferential formation of γ-crystals during nonisothermal crystallization by flow or deformation may not be correct. We also feel that our hypothesis deserves further investigation, where the in-depth knowledge would inevitably benefit our ability to control the polymorphism in iPP.

Top Priority Subjects, Zhejiang Province (Grant No. 20110903), and the U.S. team thanks the financial support by the National Science Foundation of the U.S. (DMR-0906512).

■

(1) Somani, R. H.; Hsiao, B. S.; Nogales, A.; Fruitwala, H.; Srinivas, S.; Tsou, A. H. Macromolecules 2001, 34 (17), 5902−5909. (2) Mao, Y. M.; Burger, C.; Zuo, F.; Hsiao, B. S.; Mehta, A.; Mitchell, C.; Tsou, A. H. Macromolecules 2011, 44 (3), 558−565. (3) De Rosa, C.; Auriemma, F.; Spera, C.; Talarico, G.; Tarallo, O. Macromolecules 2004, 37 (4), 1441−1454. (4) De Rosa, C.; Auriemma, F.; Circelli, T.; Waymouth, R. M. Macromolecules 2002, 35 (9), 3622−3629. (5) Busse, K.; Kressler, J.; Maier, R. D.; Scherble, J. Macromolecules 2000, 33 (23), 8775−8780. (6) Hsiao, B. S.; Wang, Z. G.; Sirota, E. B.; Agarwal, P.; Srinivas, S. Macromolecules 2000, 33 (3), 978−989. (7) De Rosa, C.; Auriemma, F.; Resconi, L. Macromolecules 2005, 38 (24), 10080−10088. (8) Seo, Y.; Hong, S. M. J. Phys. Chem. B 2007, 111 (14), 3571− 3575. (9) Kalay, G.; Bevis, M. J. J. Polym. Sci., Part B: Polym. Phys. 1997, 35 (2), 241−263. (10) Karger-Kocsis, J.; Chen, H. B.; Wu, J. S.; Varga, J. Polymer 2002, 43 (24), 6505−6514. (11) Menyhard, A.; Varga, J. Macromolecules 2007, 40 (7), 2422− 2431. (12) Auriemma, F.; De Rosa, C. Macromolecules 2006, 39 (22), 7635−7647. (13) De Rosa, C.; Auriemma, F. J. Am. Chem. Soc. 2006, 128 (34), 11024−11025. (14) Lezak, E.; Bartczak, Z.; Galeski, A. Macromolecules 2006, 39 (14), 4811−4819. (15) Chen, Y. H.; Zhong, G. J.; Wang, Y.; Li, Z. M.; Li, L. B. Macromolecules 2009, 42 (12), 4343−4348. (16) Alamo, R. G.; Kim, M. H.; Galante, M. J.; Isasi, J. R.; Mandelkern, L. Macromolecules 1999, 32 (12), 4050−4064. (17) Hosier, I. L.; Alamo, R. G.; Esteso, P.; Isasi, J. R.; Mandelkern, L. Macromolecules 2003, 36 (15), 5623−5636. (18) Kumaraswamy, G.; Issaian, A. M.; Kornfield, J. A. Macromolecules 1999, 32 (22), 7537−7547. (19) Kumaraswamy, G.; Verma, R. K.; Kornfield, J. A.; Yeh, F. J.; Hsiao, B. S. Macromolecules 2004, 37 (24), 9005−9017. (20) Somani, R. H.; Yang, L.; Hsiao, B. S.; Sun, T.; Pogodina, N. V.; Lustiger, A. Macromolecules 2005, 38 (4), 1244−1255. (21) Varga, J.; Karger-Kocsis, J. Polymer 1995, 36 (25), 4877−4881. (22) Chen, Y. H.; Mao, Y. M.; Li, Z. M.; Hsiao, B. S. Macromolecules 2010, 43 (16), 6760−6771. (23) Auriemma, F.; De Rosa, C. Macromolecules 2002, 35 (24), 9057−9068. (24) Foresta, T.; Piccarolo, S.; Goldbeck-Wood, G. Polymer 2001, 42 (3), 1167−1176. (25) Stocker, W.; Magonov, S. N.; Cantow, H. J.; Wittmann, J. C.; Lotz, B. Macromolecules 1993, 26 (22), 5915−5923. (26) Lotz, B.; Wittmann, J. C.; Lovinger, A. J. Polymer 1996, 37 (22), 4979−4992. (27) Cao, Y.; Van Horn, R. M.; Tsai, C. C.; Graham, M. J.; Jeong, K. U.; Wang, B. J.; Auriemma, F.; De Rosa, C.; Lotz, B.; Cheng, S. Z. D. Macromolecules 2009, 42 (13), 4758−4768. (28) Agarwal, P. K.; Somani, R. H.; Weng, W. Q.; Mehta, A.; Yang, L.; Ran, S. F.; Liu, L. Z.; Hsiao, B. S. Macromolecules 2003, 36 (14), 5226−5235. (29) Kalay, G.; Bevis, M. J. J. Polym. Sci., Part B: Polym. Phys. 1997, 35 (2), 265−291. (30) Wang, Y.; Pan, J. L.; Mao, Y. M.; Li, Z. M.; Li, L. B.; Hsiao, B. S. J. Phys. Chem. B 2010, 114 (20), 6806−6816. (31) Zhong, G. J.; Li, Z. M.; Li, L. B.; Mendes, E. Polymer 2007, 48 (6), 1729−1740.

■

CONCLUSIONS In situ WAXD studies of crystallization of m-iPP melt during isothermal crystallization provides direct evidence of the inhibiting or retardation effect of shear flow on the formation of γ-crystals. For the weak short-time shear sample (at strain of 300% for 30 s duration), simultaneous evolution of α- and γ-crystals occurs. The weak deformation can only increase the density of stable nuclei, but not the overall chain orientation. The relaxed chains after shear are able to crystallize into both α- or γ-crystals, and they behave similarly to the undeformed sample (PP-Static). The final fraction of γ-crystals (fγ) was 0.66 in the weak short-time sheared or undeformed sample. In this scenario, α-crystals probably served as effective seeds for nucleation of γ-crystals. In the strongly sheared samples (at strain of 500% for 30 s duration) or the long-time continuous sheared sample (at strains of 100% and 500%), the sequential occurrence of α- and γ-crystals was observed. Oriented polymer chains, constrained by the surrounding crystals after cessation of the shear and/or maintaining their extended chain confiormation by long-time shear, have a tendency to form α-crystals with the parallel chain configuration, while they do not favor the formation of γ-crystals with the nonparallel chain configuration. The emergence of γ-crystals was found to be delayed or even inhibited with the increase in shear strength, and the final fγ value was decreased substantially. This indicates that highly oriented chains have little chance to form γ-crystals directly. A simple model was used to explain the pathway for the possible formation of γ-crystals under the strong shear environment.

■

ASSOCIATED CONTENT

S Supporting Information *

Two-dimensional WAXD patterns showing the crystal growth; rheological measurements and results. This material is available free of charge via the Internet at http://pubs.acs.org.

■

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*(Z.-M.L.) Tel/Fax: +86-28-8540-6866. E-mail: [email protected]. cn. (B.S.H.) Tel: +1-631-632-7793. Fax: +1-631-632-6518. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We are indebted to Dr. Lixia Rong from Synchrotron Light Source, Brookhaven National Laboratory (U.S.A.) for the help of in situ synchrotron X-ray scattering measurement, and to Mr. Toki, from the Department of Chemistry, Stony Brook University, for refining the English language and some useful suggestions. The Chinese team thanks the financial support by the National Outstanding Youth Foundation of China (Grant No. 50925311), the National Natural Science of China (Grant No. 51033004 and 51121001), and the Open Foundation of 5062

dx.doi.org/10.1021/jp3003068 | J. Phys. Chem. B 2012, 116, 5056−5063

The Journal of Physical Chemistry B

Article

(32) Turner, J. A.; Aizlewood, J.; Beckett, D. Makromol. Chem. 1964, 75, 134−153. (33) Natta, G.; Corradini, P. Nuovo Cimento Suppl. 1960, 15, 40−51. (34) Brückner, S.; Meille, S. V. Nature 1989, 340 (6233), 455−457. (35) Dai Patrick, S.; Cebe, P.; Capel, M.; Alamo Rufina, G.; Mandelkern, L. In Scattering from Polymers; American Chemical Society: Washington, DC, 1999; Vol. 739, pp 152−165. (36) Balzano, L.; Rastogi, S.; Peters, G. W. M. Macromolecules 2009, 42 (6), 2088−2092. (37) Wang, Z. G.; Wang, R. Macromolecules 2010, 43 (23), 10096− 10106. (38) Meille, S. V.; Bruckner, S.; Porzio, W. Macromolecules 1990, 23 (18), 4114−4121.

5063

dx.doi.org/10.1021/jp3003068 | J. Phys. Chem. B 2012, 116, 5056−5063