Article pubs.acs.org/Macromolecules
Crystalline Lamellae Fragmentation during Drawing of Polypropylene Artur Rozanski* and Andrzej Galeski* Centre of Molecular and Macromolecular Studies, Polish Academy of Sciences, Sienkiewicza 112, 90-363 Lodz, Poland ABSTRACT: Filling free volume pores of the amorphous phase with the molecules of low molecular weight modifier leads to a complete elimination of the cavitation during tensile drawing. Such way of modification of a solidified material makes the polypropylene/modifier system a model system that enables the analysis of the influence of cavitation on thermomechanical properties and the mechanisms activated during its deformation. In this paper we have presented the influence of cavitation on the intensity of the lamellae fragmentation. In the case of cavitating material, on the basis of X-ray measurements and scanning electron microscopy, we have observed substantial decrease of the undisturbed crystallites lengths during its deformation up to 50−55% of their initial value. The deformation of noncavitating material proceeded with smaller decrease of average crystallites lengths by only 15−20% of their initial value. The changes of the SAXS’s long period of noncavitating polypropylene indicated that only a small fraction of lamellae stacks that are oriented parallel to the tensile direction undergo fragmentation. This type of fragmentation is connected with excessive lamellae thinning and interfacial instabilities but by no means by cavitation.
■
The first model to deal with the formation of cavities in a crystalline polymer during its deformation was a “micronecking” model proposed by Peterlin.2,12,13 Even though the model explained many phenomena taking place during plastic deformation of polymers, it contained significant inconsistency. Further research conducted in many laboratories allowed to examine and better understand the mechanisms accompanying deformation of crystalline polymers.14−18 However, the phenomenon of cavitation was only treated as an effect accompanying deformation process, related to distribution of stresses inside the polymer material, a result of operation of other deformation mechanisms. Moreover, cavitation was frequently regarded as a phenomenon which not only does not influence the course of deformation but also masks the true mechanisms activated during deformation of a crystalline polymer, thereby making it difficult to examine. Pawlak et al.19 demonstrated that cavitation of the crystalline polymers not only is responsible for material whitening but also influences the mechanical parameters such as yield stress. We have observed also substantial influence of the cavitation phenomenon on the amount of heat generated during uniaxial stretching.20 The description of the cavitation phenomenon and its influence on thermomechanical properties of semicrystalline polymers based on the papers published during the recent three decades and the newest literature data were collected in our last review.21
INTRODUCTION At a molecular scale level, yield in semicrystalline polymers1 involves the disruption of the crystalline phase in an irreversible deformation process. Upon yielding, the spherulitic structure is deformed and eventually destroyed and transformed into a fibrillar one as the plastic deformation increases.2 There are numerous other phenomena accompanying deformation of crystalline polymers. Apart from the crystalline lamella being broken into small crystalline blocks upon yielding, Liu et al. also observed that the crystalline phase is transformed into amorphous one.3 Ferreiro and Coulon4 evidenced the role of the amorphous phase on the plastic deformation at yield of a polyamide 6. Shear bands are developed in the amorphous phase originating crystalline nanoblocks, whose size increases with the strain rate. Hughes et al.5 related the onset of both intense microvoiding and stress-induced martensitic phase transitions to the yield point. Rault proposed that the yield of semicrystalline polymers involves some collective chain motions taking place in the crystalline phase.6 Nitta et al. explained the yield behavior by the disintegration of lamellar clusters (before deformed by bending due to the action of active tie molecules) accompanying lamellar fragmentation.7 Strobl et al.8−11 studied the deformation behavior of various polyethylenes under an applied tensile load based upon measurements of true stress−strain curves, elastic-recovery properties, and texture changes at different stages of the deformation process. Although Strobl et al. did not observe cavitation in their experiments and did not consider it in their analysis, cavitation was observed by others in many crystalline polymers. © XXXX American Chemical Society
Received: June 1, 2015 Revised: July 10, 2015
A
DOI: 10.1021/acs.macromol.5b01180 Macromolecules XXXX, XXX, XXX−XXX
Article
Macromolecules
In polyethylene the TEM studies26 clearly indicated that beginning at a compression ratio of 3.13, an intense fragmentation of polyethylene lamellae is taking place in plane strain compression when cavitation is prohibited due to compressive 3D stresses. In a high-magnification micrograph of ultrathin section of the 3.13 sample, these fragments of lamellae rotate in the direction away from the flow direction. This rotation of lamellar fragments is opposite to the rotation resulting from fine-chain slip. This type of behavior of stretching out of the initial lamellae and carrying the interspersed amorphous material with it, followed by their fragmentation, was reported first by Young et al.27 They have attributed the development to a so-called “coarse chain slip” in subsequent deformation, which fragmented lamellae by (100) [001] chain slip occurring in coarse steps. While this scenario appears to fit the observations at first sight, such profuse shearing should not be favored in the already highly aligned chains of the lamellae since the resolution of shear stress (Schmid factor) along the (100)[001] slip system due to the applied loading should have decreased to quite low values. Therefore, we propose a different explanation for these observations, which is more consistent with expectations of decreasing Schmid factor, and recognizing other important thermodynamic driving forces related to structural rearrangements of a topological nature and not associated with strain production. Molecules align and rotate toward deformation direction while the normals to the slip planes and the planes of the amorphous material rotate into perpendicular direction. This process continues monotonically until the lamellae have become quite elongated and have thinned down. The associated amorphous region is forced to go along with the shearing and thinning lamellae. Thin lamellae becomes unstablemuch like a layered fluid becomes unstable and responds by breakup. In fluids the stimulus for such instability of layers is the increased interface energy. In the stretching lamellae and their amorphous layers this interface instability should also be present while the required perturbations should come from thickness irregularities and other inhomogeneities. In our older paper,26 we have shown that the onset of the perturbations for a layered two-component system depends on plastic shear resistances of crystalline and amorphous layers, the interface stretching resistance, and the thickness of long-period wavelength (lamellae + amorphous layer). When this dimension becomes very small, the interface-stretching resistance becomes comparable with the plastic-stretching resistance of the crystalline and amorphous layers, and perturbations decreasing the overall interface energy are likely to grow. The dimension λ where the two resistances become equal is
In crystalline polymers the cavitation occurs preferentially in amorphous layers. It is evident that the physical parameters of the amorphous phase control the course and intensity of a cavitation process. The effect of stabilizers, additives, and low molecular weight fractions on cavitation during tensile drawing was studied in polypropylene22,23 and polyethylene.23 The additives were extracted from compression-molded samples by critical CO2 and also by a mixture of nonsolvents. The extract was an oily liquid composed of antioxidant, processing stabilizer, and a spectrum of low molecular weight fractions of polypropylene. Purified polypropylene exhibited surprisingly more intense cavitation than pristine polypropylene as it was determined by small-angle X-ray scattering and volume strain measurements. Intensification of the cavitation process in the purified samples was explained by the changes in the amorphous phase, namely the changes in free volume by eliminating low fractions and soluble additives. Increase in free volume was also confirmed by positron annihilation lifetime spectroscopy.22,23 The dominant role of the free volume of amorphous phase, which is an integral part of unordered regions of all crystalline polymers, in formation of cavitation pores proves that initiation of the phenomenon is of a homogeneous nature. It is meant that the nucleation occurs within the material itself in contrast to heterogeneous nucleation on foreign substances. That was proven by removing most of heterogeneous nuclei: impurities, additives, and gas. It appears also that in crystalline polymers a heterogeneous nucleation of cavitation is nearly inactive. More intense formation of cavitation pores in purified polypropylene proves that initiation of cavitation in polypropylene has a homogeneous nature. This anticipation is strongly supported by unusually high negative pressure necessary for cavitation for several commodity crystalline polymers at ambient conditions: polypropylene (−13.7 MPa),19 poly(methylene oxide) (−35.8 MPa),19 and polyethylene (−15.1 MPa).19 In contrast, low molecular weight substances, like tap water showing nearly −0 MPa strength, cavitate instantly. In order to show any strength against cavitation, they have to be carefully purified, and only then they can go into homogeneous nucleation range of cavitation. The presence of “empty” spaces (vapor or gas bubbles) in low molecular weight liquids is conducive to formation of cavities. The amorphous phase of polymers is characterized by a more unordered structure than amorphous low molecular weight materials due to steric hindrances introduced by chemical bonds of long chains. Hence, an intrinsic element of polymers’ amorphous phase is a fraction of so-called “free volume” resulting from its incomplete packing. At the temperatures above glass transition amorphous phase exhibits certain dynamics of free volume−“empty” spaces change in time as a result of thermal movements of polymer chains. In the papers24,25 we performed the modification of the amorphous phase of several semicrystalline polymers, cavitating during uniaxial stretching. We proved that only partial filling of the free volume pores of the amorphous phase with low molecular weight modifier leads to a decrease of intensity (polyamide 6−water system) or complete elimination of the cavitation phenomenon (polypropylene−hexane, polypropylene−chloroform, and polyethylene−chloroform systems). It appears that infusion of low molecular weight liquids into amorphous phase of crystalline polymers is a tool allowing to control cavitation and to study the influence of cavitation on deformation of crystalline lamellae in particular on their fragmentation.
λ = Χ/(cτc + (1 − c)τa)
(1)
where X is the interface energy, τc and τa are plastic shear resistances of crystalline and amorphous layers, respectively, and c is the volume fraction of the crystalline component. The estimation for deformation of polyethylene (X = 0.093 J/ m2 14,28 and τc = 7.2 MPa29 and guess τa to be around 5 MPa30 for a volume fraction of crystallites of 0.5) gives the estimate of the thickness of crystalline + amorphous layered system at the level of 15 nm. This is in the right range of long period thickness, where the breakup of lamellae is observed. Second, the breakup could be further facilitated by plastic inhomogeneities or irregularities in the thickness of lamellae. Once this process begins, the pinching-off of the lamellae could give rise B
DOI: 10.1021/acs.macromol.5b01180 Macromolecules XXXX, XXX, XXX−XXX
Article
Macromolecules
volume of the sample between markers was determined on the basis of a distance between markers and the thickness of the photographed sample. Small-Angle X-ray Scattering (SAXS). In situ SAXS studies with the use of synchrotron radiation, λ = 0.0957 nm, were performed at P03 beamline in Hasylab (Hamburg, Germany). Deformation of samples was performed on a specially designed testing machine which enabled tension of samples with a simultaneous register of SAXS scattering patterns. Symmetrically stretched samples were monitored with the use of a camera, which enabled precise calculation of local strain of the sample on the basis of change in distance between the dot markers. Tests were performed at room temperature at a standard rate 3.3 × 10−3 s−1. Deformation was conducted for 6 s, and next a scattering pattern for a given strain (calculated on the basis of images obtained with a photographic camera, with no tension) was registered; the exposure time was 12 s. The entire procedure was cyclically repeated up to the rupture of a sample. Two-dimensional scattering patterns were recorded with the use of Pilatus detector. The distance between the sample and the detector was 3000 mm. The ex situ SAXS measurements were performed with the use of 0.5 m long Kiessig-type camera equipped with a tapered capillary collimator combined with additional pinholes (300 μm in diameter) forming the beam and an imaging plate as a detector and recording medium (Kodak). The camera was coupled to a X-ray source (sealedtube, fine point Cu Kα filtered radiation, operating at 50 kV and 40 mA; Philips). The time of collection of the pattern was usually around 3 h. Exposed imaging plates were read with Phosphor Imager SI scanner and ImageQuant software (Molecular Dynamics). Wide-Angle X-ray Scattering (WAXS). The ex situ wide-angle Xray scattering (WAXS) measurements were performed with the use of photo camera. A source of Cu Kα radiation, operating at 50 kV and 35 mA, was used. Two-dimensional scattering patterns were recorded by camera equipped with a Kodak imaging plate. The distance between a sample and recording plate was 5 cm. Exposure time was approximately 10 min. Exposed imaging plates were analyzed with PhosporImager SI system (Molecular Dynamics). The diffraction of crystallographic planes (110) and (040) of PP alpha form was analyzed with the aim of determination of evolution of crystal’s dimension during uniaxial stretching, in the directions perpendicular to those planes, calculated according to the Scherrer formula:
to local strain softening and concentration of further deformation at coarser spacings. We proposed that this is the origin of the so-called “coarse chain slip” process of Young and Bowden.31 The more important matter, however, is that once the lamellae are fragmented, the fragments can change shape or giving rise to any additional lattice rotation or even rotate in the opposite direction to reduce the interface energy. Similar calculation for alpha crystals of polypropylene leads to a thickness of long period of 10.6 nm for the instability for fragmentation. In calculation the following values are used: interfacial energy (equivalent to crystal surface energy, σe), X = 0.122 J/m2,32 and similarly as for polyethylene we assumed τa = 5 MPa and crystallinity 0.5. The critical resolved shear stress for the easiest (010)[001] slip system is not known from direct measurements, but it can be deduced from the value of tensile yield stress. For polypropylene used further in this paper the tensile yield stress τy = 36 MPa.33 If the yield is activated by the easiest crystallographic slip system, then the critical resolved shear stress for the (010)[001] slip is ∼18 MPa. The main purpose of this study is to control cavitation during uniaxial stretching and to study lamellae fragmentation in cavitating and noncavitating polypropylene. The following modifiers have been chosen: chloroform and hexane. As it was presented in the papers,24,25 chosen liquids will penetrate amorphous phases of polymer but will not influence the crystalline phase. A gradual saturation of the amorphous phase with penetrants will allow to determine the impact of cavitation on intensity and habits of lamellae fragmentation.
■
EXPERIMENTAL SECTION
Materials. Studies presented in the paper have been conducted for crystalline isotactic polypropylene, whose tensile deformation is usually accompanied by cavitation. The material used was polypropylene, Borclean HB300BF (Mw = 608 kg/mol, Mn = 72 kg/mol, Mw/Mn = 8.44; manufacturer data), of melt flow index MFI = 2.5 g/10 min (for 230 °C, 2.16 kg according to ISO 1133), by Borealis. Modification. Samples for mechanical and X-ray studies were cut out from 1 mm thick film obtained by compression molding at 230 °C and cooled down between metal plates. The first batch of samples (soaked sample: SS) has been placed in a vessel containing a penetrant (chloroform or hexane) for the period of at least 72 h in order to obtain full saturation of the amorphous phase of the material with a low molecular weight liquid. The second batch of samples (reference sample: RS) provided reference material. Mechanical Testing. Mechanical properties of the materials were assessed using a testing machine (Instron 5582) of load range 0−2 kN. The shape of samples was according to the ISO 527-2 standard, with 1 mm thickness and 5 mm width. The gauge length was 25 mm. Tests were performed at room temperature at a standard rate 3.3 × 10−3 s−1. The actual shape of a sample during deformation was recorded using Nikon D50 digital camera. In order to determine the local strain, dot markers of sputter-coated gold forming a rectangular grid located in the entire gauge of a sample. Dots were at a distance of 1 mm one from another and were deposited on surfaces of the samples using an ion sputter coater and a mask obtained using photolithography. A similar measuring technique was used by us and others in the papers.24,34−36 Local strain was calculated as a change in distance between the dot markers according to the relation (l − l0)/l0, where l0 is a distance between markers for the undeformed sample and l is a distance between markers for the deformed sample. Volume strain for local strains was determined using the relation (V − V0)/V0, where V0 denotes the volume of undeformed sample. To do so, a small mirror was set up during photographic register of deformation, which directed an image of the sample’s thickness side to the digital camera. The
Lhkl = 0.9λ /(β cos θ )
(2)
where Lhkl is the crystallite length in the direction perpendicular to (hkl) plane, λ is X-ray wavelength, β is the half-width of a diffraction peak, and θ is the Bragg angle.37 To estimate the value of half-width of analyzed diffraction peaks, the 1D WAXS profiles were subsequently background-corrected and normalized. Since reflections from the different crystallographic planes frequently overlap each other, it was necessary to separate them by deconvolution. Analysis of diffraction profiles of the examined samples and peak deconvolution were performed using WAXSFit software (AHT).38,39 The software allows to approximate the shape of the peaks with a linear combination of Gauss and Lorentz or Gauss and Cauchy functions and adjusts their settings and magnitudes to the experimental curve with a “genetic” minimizing algorithm. The ex situ WAXS and SAXS patterns for SS samples were collected after chloroform removal. The removal of chloroform from samples deformed to the required strain (determined based on a distance between the dot markers) was performed in a special fixture mounted on the samples before releasing the stress from the tensile testing machine. The chloroform was removed from samples by drying at the temperature of 40 °C to constant weight. An analogous procedure was applied for RS samples in order to obtain the reference data. Differential Scanning Calorimetry (DSC). Thermal analysis of the examined materials was conducted using an indium-calibrated DSC apparatus (TA 2920, Thermal Analysis). Samples of total mass of 6−8 mg were being placed in aluminum pans and pressed slightly in order to ensure good contact with the DSC cell surface. The data were registered during heating at a constant rate of 10 °C/min, under C
DOI: 10.1021/acs.macromol.5b01180 Macromolecules XXXX, XXX, XXX−XXX
Article
Macromolecules nitrogen flow. The degree of crystallinity of the studied samples was determined according to the formula
Xc = ΔHm/ΔHm 0
affected by the state of their amorphous phase. Stress at yield point in crystalline polymers, whose amorphous phase is modified by swelling, is reduced. Swelling of amorphous phase with low molecular weight penetrant, which does not affect crystalline phase, causes the deformation of lamellae stacks. This additional tensile stress exerted on crystalline lamellae was determined by recording the stress buildup for samples with fixed ends during desorption of penetrant. It appears that measured yield stress plus the stress exerted by swollen amorphous phase amounts exactly to the stress required for plastic deformation of crystals; therefore, the yielding is determined by the same crystal plasticity despite different external load. The phenomenon was observed in several systems including polypropylene, polyethylene, and polyamide 6 with various penetrants. In order to establish the influence of the penetrant present within the amorphous phase of the polypropylene on cavitation accompanying deformation of unmodified material, small-angle X-ray scattering (SAXS) measurements were performed. Scattering patterns presented in Figures 1a,b were recorded during in situ studies using synchrotron radiation. Scattering patterns presented in Figure 1c were recorded, ex situ, after chloroform removal. The first indication evidencing the presence of cavitation during deformation of the reference material was already observed at the local strain equal to 0.08 (clearly visible at local strain 0.13). Because of the lamellar structure of the material, the cavities at this level of deformation stage are of ellipsoidal shape. Actually, their longer axis is oriented perpendicularly to the direction of deformation, which has been confirmed by the signal in the meridional region on the scattering pattern. In addition, for the sample deformed to the local strain equal to 0.3 and more, a less intensive signal appears within the equatorial area of scattering pattern. The presence of such signal at this level of deformation testifies to the formation of dilatation gaps. The characteristic feature of such discontinuities are highly deformed parts of material in the form of fibrils which create a peculiar kind of structure of the dilatation gaps. Openings occur between the neighboring fibrils, which are elongated parallel to the direction of deformation and contribute to X-ray scattering in the way shown in the mentioned scattering pattern. Microphotographs of the reference material deformed to the local strain equal to 0.5 are shown in Figure 2. The sample’s structure within polar (Figure 2a) and equatorial (Figure 2b) area of the selected spherulite is exposed. Deformation of reference polypropylene is therefore accompanied by formation of numerous cavities in polar parts of spherulites (Figure 2a) as well as individual dilatation gaps occurring mainly in the equatorial region of spherulites (Figure 2b). For higher degree of deformation of the reference sample (>0.8) we may observe an increase in the intensity of the scattering signal in the equatorial region at the expense of the signal in the meridional region. The observed change relates to reorganization of the shape of cavities, which orient themselves in the direction of the applied stress as a result of activation of subsequent plastic deformation mechanisms. For the local strain equal to 3.5 and more, significantly lengthened signal is evident within the equatorial area of scattering pattern only. At this phase of deformation, cavities undergo a significant thinning and elongation toward deformation direction. It is difficult, based on small-angle X-ray scattering measurements, to assess changes that occur in the dilatational gaps on further
(3)
where ΔHm is the measured specific heat of melting and ΔHm is the heat of fusion of alpha polypropylene crystals. For polypropylene the value of ΔHm0 = 209 J/g has been assumed.40 The net thermal signal from the polypropylene mesophase was obtained by integration of the resulting endotherm below the baseline in the temperature range from +50 to +100 °C without splitting between melting and recrystallization. The baseline was plotted as a sigmoid between 30 and 190 °C using the TA Instruments software. The estimated uncertainty of enthalpy determination is ±0.5 J/g. Scanning Electron Microscopy (SEM). The bulk morphology of the samples prior to deformation and samples deformed to selected strains was studied using a scanning electron microscope (SEM, Jeol JSM 5500LV). The internal part of a sample was exposed by cutting with an ultramicrotome (Power Tome XL, Boeckeler Instruments, Inc.) equipped with a diamond knife (Diatome Ltd.). The exposed surfaces were etched for 2 h at room temperature in a solution composed of KMnO4 (0.7 wt %) dissolved in a mixture of concentrated sulfuric acid, orthophosphoric acid, and distilled water in the volume ratio 5/4/1, in accordance with the procedure proposed by Basset.41 To improve etching, the mixture was placed in an ultrasonic bath running periodically for short time periods during the etching process (for approximately 2 min every 20 min). After completion of etching, the samples were immersed into four tubes in the following order: with diluted sulfuric acid (sulfuric acid/water 2/7 vol), perhydrol, distilled water, and acetone. Washing was run in the presence of ultrasound, so as to ensure more efficient transport from/ to the surface of etching remnants/liquids. Dried sample was coated with a fine gold layer (about 20 nm) by ion sputtering (JEOL JFC1200) and examined under a scanning electron microscope (JEOL JSM-5500 LV). Microphotographs were registered in a high-vacuum mode at the accelerating voltage of 10 kV. Microscopic image was created using secondary electron detector (SEI). Transmission Electron Microscopy (TEM). The lamellar structure of selected samples was examined with a transmission electron microscope (Tesla BS 500, Tesla, Czech Republic), operating at the accelerating voltage of 90 kV. Samples for TEM observations, in the form of ultrathin sections 60 nm thick, were prepared by cryoultramicrotoming. The cryo-ultramicrotome (PowerTome PC, Boeckeler, USA) equipped with a 35° diamond knife (Diatome, Switzerland) was used for sectioning. Before cutting polypropylene samples were exposed to the vapor of RuO4 at room temperature for 24 h. 0
■
RESULTS Modification of the amorphous phase of polypropylene with chloroform or hexane led to very similar results; hence, only the detailed results obtained for the polypropylene modified with chloroform will be presented further. In order to estimate the amount of chloroform present in the examined material on completion of the conditioning process, TGA examination has been conducted in air. On the basis of the thermogram the amount of chloroform has been estimated at 16.5 wt % (11.3 vol %) in the analyzed material on completion of the conditioning process. Desorption of penetrant under laboratory conditions was found to be a relatively quick process; therefore, all measurements were performed within a minute after removal of the sample from the conditioning vessel. The stress−strain curves of both materials (reference and chloroform-soaked) are typical; however, the yield stresses differ. In our previous paper42 the stress−strain behavior is described in detail. In particular, it is shown that plastic yielding of polypropylene and other semicrystalline polymers is greatly D
DOI: 10.1021/acs.macromol.5b01180 Macromolecules XXXX, XXX, XXX−XXX
Article
Macromolecules
(>0.8) stages of deformation; however, microphotographs presented later in this paper indicate a rapid increase in their size combined with the reorganization of shape and destruction of fibrillar structure. Such changes in the morphology of the material are outside detection of the SAXS method. Scattering patterns for chloroform-soaked polypropylene, for the same degree of local strain, were also collected and presented in Figure 1. Up to the local strain equal 4.5, the scattering patterns do not display any signals characteristic for cavitation. Conclusions supported by X-ray examinations, indicating appearance of cavitation in reference samples and excluding the phenomenon in the modified polypropylene samples, are also confirmed by the results of volume strain measurements. Deformation of samples of unmodified polypropylene is accompanied by a significant volume increase by approximately 30%, an effect of formation of cavities and dilatation gaps. Filling the free volume of the amorphous phase of polypropylene with a low molecular weight penetrant eliminates the process of formation of discontinuities in the material, which is manifested by the lack of volume change of samples during their deformation. Filling the pores of free volume of the amorphous phase of polypropylene results in elimination of cavitation, which together with the lack of changes in the crystalline regions, as demonstrated in the paper,24 provides the model system for analysis of the influence of cavitation on crystalline lamellae fragmentation during drawing. Appearance of cavitation, and change in the shape of cavities in particular, must be accompanied by a change in the arrangement of crystals in the material. Specific organization of crystallites at various stages of deformation can be examined using wide-angle X-ray diffraction (2D WAXS). Diffraction patterns recorded for subsequent deformation stages of samples of unmodified polypropylene and saturated with a low molecular weight penetrant are presented in Figure 3. The concentric rings represents diffraction from the following crystallographic planes: (110), (040), (130), and (111) together with (−131) and (041). For the deformed samples, the diffraction patterns up to the local strain of 0.5 do not exhibit significant degree of orientation of crystallites in the examined materials. The orientation of lamellae within spherulites in the drawing direction is already highly advanced at the strain of 1.2 (Figure 1c). However, below the strain 0.5, in contrast to high lamellae orientation, the 2D WAXS patterns in Figure 3 show still circular diffraction rings from weakly oriented crystalline planes, indicating far lower crystal
Figure 1. In situ small-angle X-ray scattering patterns of a series of polypropylene samples: (a) reference sample-RS; (b) chloroform soaked sample-SS; (c) ex situ SAXS patterns of SS samples after removal of chloroform. Chlorine in the modifier was masking the scattering from lamellar arrangement and had to be removed. The samples during removal of chloroform were kept with fixed ends in a special frame in order to prevent strain recovery. Numbers correspond to local strain of samples. Direction of deformation: vertical.
Figure 2. Microphotographs of unmodified polypropylene sample deformed to the local strain of 0.5. Morphological details unraveled by cryomicrotoming of the skin layer followed by etching. Direction of deformation: vertical. (a) Polar region of spherulite; (b) equatorial region of spherulite. E
DOI: 10.1021/acs.macromol.5b01180 Macromolecules XXXX, XXX, XXX−XXX
Article
Macromolecules
Figure 3. Ex situ WAXS patterns for samples: (a) RS and (b) SS after removal of penetrant. Numbers correspond to local strain of samples. The samples during removal of chloroform were kept with fixed ends in a special frame in order to prevent strain recovery. Direction of deformation: vertical.
planes (overlapping reflections from (111), (−131) and (041) at around 2Θ = 22°) will form first order layer of X-ray fiber diffraction pattern because the last number in crystallographic planes indices is 1. Changes in the arrangement of crystallites during deformation, both for RS samples (with cavitation) and SS (without cavitation), run similarly: the zero and first order layers are formed in patterns for both samples. The essential difference is the broadening of individual X-ray reflections for RS samples so, they strongly overlap, while the diffraction patterns recorded for SS samples remain sharp up to the local strain of 4.5. Similar broadening and overlapping of (110), (040), and (130) X-ray reflections as for our RS sample was observed in the past by Hsiao et al.44 for deformed PP fibers. They considered two possible explanations: deformed crystals may be defective, or there is the presence of mesomorphic form mixed with the α-crystals. Their results indicated that defective α-form crystals were present in the initial iPP fibers. These αform crystals were completely converted into the mesomorphic form at draw ratios above 2.5 at room temperature. However, the above conclusion was based on an image analysis method to deconvolute the quantitative information on the mass fractions of the crystalline, mesomorphic, and amorphous phases from the 2D WAXD patterns. In our opinion such analysis is erroneous because in 2D WAXS patterns only those crystals are considered which orientation fulfills the Braggs rule. Other crystals are oriented at other than Braggs angles are not detected and escape the analysis. This is especially true for the amorphous phase being oriented by shear at 45° with respect to fiber axis. Hence, most of the oriented amorphous phase is not detected by 2D WAXS. The correct route to obtain the
orientation than the orientation of lamellae. A similar phenomenon was observed by us in polypropylene deformed without cavitation at elevated temperature43 for samples that, upon reaching the required strain, were relaxed with fixed ends for several hours. The explanation for this phenomenon was as follows: the rotation of lamellae toward drawing direction is associated with the reverse rotation of chains in crystals due to fine chain slips. These two rotations in opposite directions counterbalance resulting in a much weaker crystallographic orientation than expected from the ex situ SAXS patterns (Figure 1c) and SEM images (Figure 13). Such rotations occur also when cavitation and lamellae fragmentation are in action during drawing. For the higher deformed samples (>0.5), of both unmodified and modified polypropylene, one observes concentration of signals in the selected regions of diffraction circles. Planes containing the axis of macromolecular chain [(110) (040) (130)] tend to orient themselves parallel to the direction of the applied stress (diffraction signal in the equatorial area). Meanwhile, the external circle, appearing as a result of overlapping of signals from three different crystallographic planes, all with the index l = 1, not containing axes of macromolecular chains, transforms into a four-point pattern. This indicates the presence of two populations of lamellae in the material. As the strain increases reflections from (110), (040) and (130) planes concentrate in equatorial zone of diffraction pattern forming so-called “zero layer” of X-ray fiber diffraction pattern. For pure uniaxial straining the intensities of reflections from all crystallographic planes containing macromolecular chains should increase in a similar way. Reflections from other F
DOI: 10.1021/acs.macromol.5b01180 Macromolecules XXXX, XXX, XXX−XXX
Article
Macromolecules
amorphous background with strongly broadened reflections from the crystallographic planes. Diffraction profiles obtained for samples saturated with a low molecular weight penetrant for subsequent deformation stages are presented in Figure 5. As in the case of reference material, a clear increase in the intensity of diffraction profile is observed in the samples with higher local strain, an effect of the concentration of reflections from crystallographic planes of indices (hk0) or (0k0) in the horizontal region of scattering images. However, one does not observe a relative reduction in the intensity of peaks from the crystalline component in relation to the amorphous background. Reflections from these planes remain sharp up to the local strain of 4.5. Figures 6 and 7 present DSC thermograms recorded for the reference material and saturated with chloroform deformed to Figure 4. Horizontally scanned diffraction profiles of WAXS scattering patterns for reference samples deformed to local strains depicted on the graph.
Figure 6. DSC thermograms of a series of polypropylene samples (reference material). Numbers correspond to local strain of samples. Figure 5. Horizontally scanned diffraction profiles of WAXS scattering patterns for chloroform-soaked polypropylene samples. The samples during removal of chloroform were kept with fixed ends in a special frame in order to prevent strain recovery.
information on the degree of crystallinities of oriented samples is by integration of total intensities under rectified and deconvoluted diffraction peaks for certain crystallographic forms and for the amorphous halo over all positions of the sample for Euler angles α from 0 to 90″ and β from 0 to 360″ and scaled to the integrated total intensity under all diffraction peaks, with the amorphous halo background being subtracted. Such a procedure was described in detail and successfully applied by Galeski et al.45 Figure 4 presents equatorial diffraction profiles of the scattering images from Figure 3, for reference material, scanned for individual local strains. In the case of cavitating samples an increase in the intensity of the obtained diffraction profile is observed on subsequent stages of deformation accompanied by a gradual broadening of the reflections from individual crystallographic planes. The intensity ratio of signals from the crystalline component to the intensity of amorphous background gradually decreases or the mean width of reflections from crystalline component gradually increases. Therefore, for the reference sample of local strain of 4.5 one observes only a broadened signal as a result of overlapping of a peak of the
Figure 7. DSC thermograms of a series of polypropylene samples (chloroform-soaked material). Numbers correspond to local strain of samples. Chloroform was removed before DSC scans.
selected strains. Because of numerous phenomena and mechanisms activated during deformation of both the reference and the noncavitating samples (lamellar distortion, orientation of crystallites, defects in crystals caused by the presence of dislocations, reorganization of the crystalline structure induced by deformation of the material: monoclinic form−mezo phase), G
DOI: 10.1021/acs.macromol.5b01180 Macromolecules XXXX, XXX, XXX−XXX
Article
Macromolecules
Figure 8. Change of the value of degree of crystallinity and net melting enthalpy of mesophase for reference and chloroform-soaked samples as a function of local strain. Open symbols for reference cavitating RS samples and filled symbols for noncavitating SS samples.
Figure 9. Exemplary results of the peak separation applied to the reference polypropylene deformed to the local strains: (a) 0 and (b) 1.4.
noncavitating polypropylene for varied local strains is presented in Figure 8. In both cases of the cavitating and noncavitating materials the degree of crystallinity is not substantially changed for increasing local strain. It oscillates between 52 and 56 wt %. Also, the content of a mesophase at different values of local strain seems to be at similar level in the case of RS and SS samples. Strong change in the diffraction profile of cavitating samples with no significant change in the degree of crystallinity (Figure 8 and DSC thermograms discussed above) suggests strong reduction in undisturbed sizes of crystals. The lamellae thinning and fragmentation may be responsible for such effect. For WAXS profiles recorded for varied local strains (Figure 4 and Figure 5) we determined the values of the unperturbed crystallite length in the direction normal to the population of crystallographic planes (110) and (040) based on the Scherrer equation. Before determining the half-width of the signals coming from the above-mentioned crystallographic planes, proper diffraction profiles have been deconvoluted with the use of WAXSFit software in a way presented in Figure 9. It must be noticed that the mesophase concentration was low in all successive deconvolutions.
and relaxation of orientation during heating, interpretation of changes in the shape and location of the melting peak is very difficult. Analysis of thermograms in a low temperature region (
