Article pubs.acs.org/Macromolecules
In Situ Observations of Flow-Induced Precursors during Shear Flow Yunfeng Zhao, Kouhei Hayasaka, Go Matsuba,* and Hiroshi Ito Department of Polymer Science and Engineering, Yamagata University, 4-3-16, Jonan, Yonezawa, Yamagata 9928510, Japan ABSTRACT: In this work, the formation of flow-induced precursors (FIPs) during and immediately af ter applying a shear flow in isotactic polypropylene (iPP) melt is studied at high spatial resolution and high time resolution with a high-speed polarized optical microscope (HSPOM) and simultaneous small/wide-angle X-ray scatterings (SAXS/WAXS). A number of studies have reported the formation of FIPs in the early stage of flow-induced crystallization. However, reports on direct observations of the formation of precursors during shearing are scarce. We find that FIPs are observed using HSPOM after a certain incubation time during shearing. After cessation of flow, some FIPs dissolve gradually and others transfer to crystals. This suggests a competition between chain relaxation and crystallization. In SAXS measurements, the formation of FIPs during shearing is indicated by weak equatorial streaks normal to the flow direction after a specific incubation time. The intensity integrated from 2D SAXS images shows a sudden decrease upon stopping the shear flow before it increases again during isothermal crystallization, similar to the results in rheo-optical measurements. WAXS measurements capture extremely low signals of crystallinity in the primary stage of the formation of FIPs, which is indicative of the very small magnitude of crystals in FIPs. Three crystalline components (alpha, beta, and gamma form) separated by Lorentzian fittings emerge in the late stage of shearing. The crystallinity of all components decreases immediately after shearing, and the fractions of the beta and gamma forms also decrease, which suggests a faster dissolution of the beta and gamma forms compared with that of the alpha form. Finally, residual precursors lead to the wellestablished “shish-kebab” morphology.
■
INTRODUCTION Semicrystalline polymers are now a widespread class of basic materials with a variety of applications. The internal characteristics of polymers such as mechanical and optical properties are strongly affected by the degree of crystallinity (e.g., the crystalline component imparts strength, whereas the amorphous component imparts flexibility). The processing procedures such as extrusion, injection, and spinning are necessary during the fabrication of polymer products, which have a dramatic impact on final morphology and properties. Different from the spherulitic morphology, in which there is a 3D random growth of folded chain crystals (FCCs), when polymers are subjected to external fields such as shear and elongation, hierarchically the subnano- to microscale crystalline structures demonstrate a distinct scenario.1−8 Nucleation as well as crystal growth is enhanced in flow-induced crystallization (FIC), as reported in numerous studies since the 1960s.9−14 The morphology alters that of the well-known “shish-kebabs”, which are composed of the highly extended long filament (shish) as a backbone on which disk-like fold chain lamellae (kebab) grow epitaxially.15−18 However, the nature of shish-kebabs, especially the structural formation in a very early stage of flow-induced crystallization, has not been well-understood because of its nonequilibrium property. Recently, with advanced characterization techniques such as time-resolved rheo-X-ray/neutron scattering, several authors19−27 have obtained some new insights into flow-induced © 2012 American Chemical Society
precursors (FIPs), which could be a key to clarify the molecular origin of shish-kebabs. Somani et al.19,20 studied FIPs in an isotactic polypropylene (iPP) melt near its nominal melting temperature. Oriented structures with a scale of hundreds of angstroms were detected immediately after cessation of shear, but they did not show any crystalline reflection. The effect of molecular architecture on the nature of FIPs was investigated by Agarwal et al.21 in a long-chain branched isotactic polypropylene (LCB-iPP) melt. The fraction of FIPs in LCBiPP was enhanced compared with linear iPP. Li et al.22,23 demonstrated that smectic ordering bundles emerged prior to crystallization in a supercooled melt. These initial smectic bundles provided nucleation sites similar to shish and induced the epitaxial growth of kebabs. The first in situ evidence of the formation of FIPs during shearing was presented by Balzano et al.24 using X-ray scattering. FIPs were generated and already crystallized during shearing at a probable temperature and sufficiently intense shear flow. Our previous work25,26 and that of Alfonso et al.27,28 demonstrated flow-induced macroscale precursors in isotactic polystyrene (iPS) at temperatures above the nominal melting temperature after applying a short-term shear flow. These macro-FIPs had a very long lifetime (primarily dependent on Received: September 10, 2012 Revised: November 17, 2012 Published: December 13, 2012 172
dx.doi.org/10.1021/ma301888r | Macromolecules 2013, 46, 172−178
Macromolecules
Article
temperature, flow, molecular weight, etc.), although initially they dissolved partially after cessation of flow. However, the nature of these precursors from the nano- to microscale was still unclear despite considerable efforts. Moreover, the structural formation (i.e., FIPs) during shearing has been less studied because of the fast dynamics of FIPs, which necessitates high time resolution for precise in situ observations. In this study, we performed a series of measurements involving high spatial resolution with a high-speed polarized microscope and synchrotron X-ray scattering with a fast imaging detector to probe the formation process of FIPs during and immediately after shearing.
■
Figure 1. Temperature protocol of shear experiments for rheo-SAXS/ WAXS and rheo-optical measurements.
EXPERIMENTAL SECTION
Materials. The polymer used in this study was a commercial grade iPP purchased from Prime Polymer (Tokyo, Japan) with Mw = 3.0 × 105 and Mn = 7.2 × 104, where Mw and Mn are the weight-average and number-average molecular weight, respectively. The tacticity of iPP was 90.5%. The nominal melting temperature (Tm) of iPP was 160 °C in differential scanning calorimetry (DSC) measurements with a heating rate of 10 °C/min. Characterization Techniques. Small Amplitude Oscillatory Shear (SAOS) Measurements. The viscoelastic properties of iPP melts at 148 °C were performed with a TA Instruments (Tokyo, Japan) parallel disk type rheometer ARES G2. Samples with a geometry of 25 mm in diameter and 1 mm in thickness were measured at a small strain amplitude γ = 1%. The angular frequency range was 0.1−100 rad/s. A nitrogen atmosphere was applied to samples in order to prevent degradation. Rheo-SAXS/WAXS and Rheo-Optical Measurements. Timeresolved simultaneous small/wide-angle X-ray scattering (SAXS/ WAXS) measurements were performed using an apparatus installed at the beamline BL6A, Photon Factory, KEK (Tsukuba, Japan). A 2D CCD camera (C7300: Hamamatsu Photonics K.K., Hamamatsu, Japan) with an image intensifier (resolution = 1280 × 1024 pixels, pixel size =125 × 125 μm) was used as the SAXS detector. For WAXS measurements, a flat-panel type detector (C9728DK: Hamamatsu Photonics K.K.) with a resolution of 1032 × 1032 pixels and a pixel size of 50 × 50 μm was applied. The wavelength λ of the X-ray beam was 1.5 Å. The camera length was 2.3 m for SAXS and 60 mm for WAXS measurements. Because of the fast dynamics of FIPs, the frame rate for SAXS/WAXS measurements was set to 2 images/s. High-speed polarized optical microscope (HSPOM) measurements were performed with a modified Keyence VW-5000 microscope (Osaka, Japan) equipped with a high-speed CCD camera attachment (maximum frame rate = 24 000 images/s). A Linkam CSS-450 hightemperature shear cell (Surrey, U.K.) was used to control the temperature and shear conditions. Samples 50 μm thick were placed between two quartz plates in HSPOM measurements and between two stainless plates with windows of Kapton in SAXS/WAXS measurements after temperature calibration. The temperature protocol for shear experiments is shown in Figure 1: samples were (a) heated to Tmelt (Tmelt = 210 °C for iPP, around its equilibrium melting temperature (Tm0)29) from room temperature at a rate of 30 °C/min; (b) held at Tmelt for 5 min to erase thermal history; (c) cooled to the shear temperature (Tshear) at a rate of 30 °C/min; and (d) held at Tshear for 3 min to reach the temperature equilibrium and then subjected to a shear flow (100 s−1 for 20 s); and (e) after cessation of shear flow, iPP samples were quenched to a desired temperature (Tc) for isothermal crystallization (in this study, Tshear = Tc = 148 °C).
■
the product of shear rate and relaxation time (i.e., reptation time τrep for chain orientation and Rouse time τR for chain stretch) of the longest molecules, is widely used to estimate the effect of flow on the molecular conformations.37−40 The Derep (for chain orientation) and DeR (for chain stretch) thus depend on the shear rate. Three cases can be considered for an entangled polymeric system.39−41 (i) Derep < 1 and DeR < 1: reptation42−44 and contour length fluctuation (CLF)45 are the dominant relaxation modes and chain segments relax with the slowest time scale τrep (reptation time), suggesting that the flow intensity is not sufficiently strong to induce chain orientation/ stretch; (ii) Derep > 1 and DeR < 1: convective constraint release (CCR)46 becomes the dominant mode, which indicates the disentanglement of the polymeric topological network; and (iii) Derep > 1 and DeR > 1: chain stretch between entanglements. Therefore, FIPs induced at a moderate flow intensity (Derep > 1 and DeR < 1) showed little electronic density difference from amorphous chains because polymer chains are oriented but not stretched.25 To clarify FIPs and enable their detection in SAXS/WAXS measurements, a large flow intensity (Derep > 1 and DeR > 1) is required. De numbers at 148 °C were estimated by fitting the SAOS data (Figure 2) with a multimode Maxwell mode47 N
G′(ω) =
∑ i=1
Gi(ωτi)2 [1 + (ωτi)2 ]
(1)
RESULTS AND DISCUSSION
Shear flow always promotes nucleation (nucleation rate/ density) and thus accelerates the onset of crystal growth (secondary nucleation) at a sufficient intensity.30−36 The Deborah number (De)/Weissenberg number (Wi), defined as
Figure 2. Storage (○) and loss (□) moduli of iPP melts at 148 °C fitted by Maxwell mode (solid lines). 173
dx.doi.org/10.1021/ma301888r | Macromolecules 2013, 46, 172−178
Macromolecules
Article
Figure 3. Time-resolved micrographs of iPP melts during shearing. The shear temperature (Tshear) is 148 °C. The shear rate is 100 s−1. The arrow indicates the flow direction.
Figure 4. Time-resolved micrographs of iPP melts during isothermal crystallization (Tshear = Tc = 148 °C) after cessation of flow. Two characteristic FIPs labeled as A and B are used to illustrate the relaxation and crystallization of FIPs. N
G″(ω) =
∑ i=1
Janeschitz-Kriegl et al.32 examined the effect of shear time on the number of nuclei by calculating the work applied to the unit volume of the sample. First, HSPOM observations were performed to investigate the formation of macro-FIPs during/after shearing. The iPP melts were subjected to shear flow with a shear rate of 100 s−1 for 20 s when the temperature was quenched to 148 °C (Tshear = Tc = 148 °C). Such a shear rate was thought to be sufficiently intense to induce FIPs, which could be characterized by SAXS/ WAXS as reported (Derep > 1 and DeR > 1).24 The shear time of 20 s applied here was longer than the previously used “shortterm” shear time (usually
