Morphological analyses evidencing corrugate-grating lamellae assembly in banded spherulites of Poly(ethylene adipate)

Highlights

• Microbeam X-ray scattering and interior morphology prove corrugate-grating assembly inside PEA ring-banded spherulites.

• 3D interior proves discontinuous interfaces between the tangential and radial lamellae with perpendicular intersection.

• The grating structure, not continuous helix-twist classical model, fits with experimental results of assembly periodicity.

Abstract

The classical model of continuous helix-twist lamellae versus a novel corrugate-grating assembly in accounting for long-debated issues of periodic optical banding in poly(ethylene adipate) (PEA) polymer spherulites is critically assessed, by comparing and checking against a series of scientific evidence based on experimental data of scanning electron microscopy (SEM), ultrathin microbeam small-angle & wide-angle X-ray scattering (SAXS/WAXD), and optical microscopy. Analyses show that grating-structured model fits perfectly with all results. The novel proposition of grating-structured lamellae with a crossbar pitch fits perfectly with the experimental optical band spacing, and the interior morphology dissection is fully supported by the literature-reported microbeam X-ray data.

Introduction

Poly(ethylene adipate) (PEA) ring-banded spherulites (RBS), like classical polyethylene (PE) ring bands, are widely studied and reported in the literature [[1], [2], [3], [4], [5], [6], [7], [8], [9], [10]]. However, the mechanism of lamellar twisting is still yet to be fully understand, and need views from alternative approaches for fuller and better understanding. As a matter of fact, random twists in lamellae can be found in lamellae of virtually any type of polymer spherulites – regardless bands or not [11,12]. For the “continuous lamellar twist model” being responsible for optical bands in spherulites, it requires the “continuity” and “synchronism” of thousands of lamellae to twist in perfect pace to achieve a clear optical band, which is yet to be proven without doubt. In probing the mechanisms for concentric optical bands in polymer spherulites, it is not enough to prove that occasional twists by sporadic happenstance are present in lamellae, but more critical to provide two pieces of information that until today are yet to be established in the literature. There are two missing pieces of morphology evidence for justifying the helix-twist lamellae being responsible for the optical banding in polymer spherulites: (I) band spacing in POM micrographs on banded spherulites cast in thin-film polymers must be comparable with the half-pitch of helicoid lamellae presented as evidence; and (II) valid SEM/TEM evidence for presence of long helicoid lamellae in banded spherulites. However, these two critical pieces of evidence are yet to be provided beyond doubt. A recent review article [13] by Woo et al. emphatically pointed out that although there have been numerous reports in past 60 years attempting to prove that polymer lamellae twist in banded spherulites, like those many demonstrated occurrences in small-molecule compounds, there has been no direct morphology evidence showing the lamellae's screw-like pitches matching exactly with the optical inter-ring spacing. There are always huge discrepancies between the size of helix-pitch of twisted lamella (ranging a few hundred nanometers) in TEM evidence and the optical band spacing (ranging a few micrometers to 30 μm). The tens or hundreds times of discrepancies in the twist pitch scales vs. optical band spacing could not be plausible explained, and has not been satisfactorily settled, as shown in various investigations reporting TEM/SEM morphology of helix-twist lamellae [[14], [15], [16], [17]]. A twist pitch in nanometer scales can hardly be justified for the tens of micrometer optical banding commonly seen in spherulites. This experimental fact means that the lamellae twist (random helix or irregular scrolls/bends) may be a habit of crystal growth in a compactly squeezed space that is not necessarily limited to occur only in banded spherulites, but also in ringless ones. In addition, many investigators used solvent etching to treat specimens of banded spherulites in cast thin films, which have recently been reported to induce artificial twisting by solvent exposures [[18], [19], [20], [21], [22]].

X-ray is one of the most fundamental tools for analyzing crystals and their assembly. It is a natural instinct that investigators in past decades conceived a handy model: if a micro-size X-ray beam could be directed and focused onto specific spots of the assumed model of “continuous helix-twist lamella” with stepwise motion (spaced by ~ 1 μm), then in principle, the assumed crystal structure of continuous helix-twist could be proven by interpreting the signals obtained for each discrete spots on the helicoid lamellae. This thought may have been over-simplified and several key points overlooked, as we will expound later. A very first but rudimentary “microbeam” in X-ray was conceived 50 years ago; nevertheless, the intensity was too low and beam size was too big for the intended purposes. Fujiwara [2] in 1960 used a then rudimentary tool of pin-holed lead (square hole of 2.5 μm × 2.5 μm) to block an X-ray beam (generated by conventional X-ray tube) to design a “microbeam” in trying to analyze the then-proposed helix-twist lamellae in crystallized PE bands. Apparently, the actual X-ray beam size in Fujiwara's early study would be too large, and extremely too low in intensity, to deal with the 4 μm pitch in banded PE spherulites. The helix-twist K–P model was further reinstated by Lotz and Cheng [6], and Toda et al. [23], among many others not all cited, in later years. After fifty years, the powerful synchrotron X-ray sources, coupled with Frenel-zone plate (FZP) for beam focusing, have made possible ultrathin X-ray beam approaching sub-microns in beam diameter. Rosenthal et al. [24,25] in 2012 used such an advanced synchrotron microbeam [using ID13 beamline of the European Synchrotron Radiation Facility (ESRF) (Grenoble, France)] to study lamellae in banded spherulites of poly(trimethylene terephthalate) (PTT) as a “re-visit” to Keith-Padden's (K–P) helix-twist model [8,[26], [27], [28]] in 1964. However, Rosenthal et al., in conclusion of their study [24,25], frustratingly stated: “Despite the formal agreement with the K–P model, the value of the chain tilt of 4° alone is too faint to be the sources for generation of the surface stresses required for twisted lamellar growth”. Apparently, even with much more advanced and powerful modern instruments than those available in 1960's, solid proofs on “continuity” and “source for twisting” are yet still controversially debated to date, which suggests that critical pieces of evidence in support of the twist proposals are yet to be discovered. Similarly, in 2013, Schultz [29] claimed in a re-visit of his own old 2004 AFM data [30] for poly(3-hydroxybutyrate-co-3-hydroxyhexanoate) (PHBH) ring-banded spherulites that he emphatically reverted the opinion of continuous lamellae twisting to “unusual discontinuity” existing in the bands. He also emphasized “what observed here does not accord with any extant models of crystal twisting.” As long-year progress in proposition on the model could not be consistently or universally made in past 60 years, it is worthy and necessary to probe if there might be other more plausible mechanisms.

As completely novel proposals for banded spherulites, several updated studies have clearly revealed the formation of mutually perpendicular orientations of crystal lamellae in PEA with the layer thickness (revealed in SEM interior dissections) being equal exactly to the optical pitch in POM micrographs and its relation to the alternate optical birefringence in PEA ring-banded spherulites [3,13,[31], [32], [33], [34]]. A grating-like structure in the banded PEA spherulites is proven in a recent work, which convinces that the crossbar pitch (equivalent to the inter-flute spacing in corrugate-board structure) is exactly equal to the optical band spacing (6.7 μm) [35]. Three pieces of critical evidence were presented for the proposed mechanisms: (I) there exists distinct discontinuity between the successive bands owing to interfaces in tangential-to-radial crystal transitions, (II) X-ray signals of ultra-thin microbeam, being narrow enough, for the tangential crystals could be feasibly compared to those for the radial ones, and (III) microbeam X-ray signals in the tangential crystals (of about 1–2 μm width) are abruptly different from those in the radial crystals (of about 4–5 μm). These facts suggest that there is no continuity in going from the tangential crystals to radial ones.

Moreover, the lamellar architecture of PEA banded spherulite was amply exemplified by SEM morphology analyses. It would be ideal if the morphology of the novel proposition could be further justified by data using an ultra-thin microbeam (beam size narrowed to ca. 1 μm) in order to give a manifest proof about the mechanisms of perpendicular corrugate-board structure in ring-banded spherulites. By utilizing the morphological results of interior-dissected PEA banded spherulites, which were compared with published data of ultrathin microbeam SAXS and WAXD data (from synchrotron sources) taken from cited literature, such as those by Rosenthal et al. [24] or Tashiro et al. [36], this work aimed to further justify a novel interior assembly to fully prove the validity of a grating structure, and not continuous lamellae helix-twisting, is at work to produce the well-known periodicity in optical rings. This work further utilized 3D approaches to prove beyond doubts that a crossbar grating structure is sufficient to resolve the long-debated origins of optical periodic bands in crystallized polymers. In addition, the dissected interior crystal assembly of PEA could be justified with literature data of microbeam analyses to fully reinforce the experimental SEM characterization results on the interior morphology of banded polymer spherulites.