Two-photon optical tomography of dye- and nanoparticle-labeled iPP and PLA revealed unexpected three-dimensional crystallization morphologies — nonspherical spherulites, 'vases' and 'goblets', mirrored spherulite nucleation across films, shish-kebab structures grown from quiescent melt, and accumulation of 200 nm silica nanoparticles at spherulite boundaries — and attributed them qualitatively to a crystallization-induced negative pressure in the constrained melt. What remains open is the quantitative physics: the force balance that rejects and transports particles far too large to diffuse away (Stokes–Einstein diffusion is ~1000× too slow), the predictive link between negative pressure, melt flow, nucleation kinetics and cavitation, the dependence of these morphologies on molecular weight, polymer chemistry, film thickness and nanoparticle properties.

Background

Semicrystalline polymers such as isotactic polypropylene (iPP) and poly(lactic acid) (PLA) solidify from the melt into spherulites — spherical, radially growing aggregates of lamellar crystals — and their morphology on the 1–1000 µm scale governs mechanical and transport properties. Shish-kebabs (row-nucleated structures: a central extended-chain fiber, the 'shish', decorated with chain-folded lamellar plates, the 'kebabs') normally require flow or external stress to form. Because crystalline polymer is denser than the melt, crystallization causes volume contraction; in a film confined between glass plates this contraction cannot relax freely, producing a negative pressure (tensile stress) in the remaining melt, which can drive melt flow, cavitation (void formation), and even acoustic emission — an effect known indirectly since Galeski, Koenczoel, Piorkowska and Baer measured acoustic emission during polymer crystallization (Nature 325, 40–41, 1987, DOI: 10.1038/325040a0) and Pawlak and Galeski analyzed spherulite growth-rate stability under this pressure drop (J. Polym. Sci. B 28, 1813–1821, 1990, DOI: 10.1002/polb.1990.090281012).

Yang, Wei, Cseh, Kazemi, Zeng, Xie, Saba and Ungar (Nat. Commun. 12, 5054, 2021, DOI: 10.1038/s41467-021-25297-w) visualized these phenomena directly in 3D for the first time by two-photon confocal fluorescence tomography — an optical sectioning technique in which a femtosecond pulsed laser excites a fluorescent dye (Nile Red) only at the focal point, allowing deep, low-photodamage 3D imaging of where dye, crystalline polymer, and labeled 200 nm silica nanoparticles (NPs) reside in iPP and PLA films crystallized between glass slides. Their 2D precursor study (Yang, Xie, Saba, Cseh, Ungar, Polymer 191, 122246, 2020) had already shown that dye and even large NPs are rejected by growing spherulites and pile up at growth fronts, contrary to the expectations of the Stokes–Einstein relation (the standard formula D=kBT/(6πηr)D = k_BT/(6\pi\eta r) for the diffusion coefficient of a sphere of radius rr in a medium of viscosity η\eta) — the estimated diffusion rate of a 200 nm NP in the melt is about a thousand times too low for the particle to escape a growing spherulite. The 2021 paper added: nonspherical spherulites elongated toward the confining glass (up to 25% vertical diameter excess in PLA), 'vase' and 'goblet' cavities, spherulite nucleation mirrored almost exactly across a film (top-surface spherulites nucleating within seconds of those on the bottom, communicating via pressure-driven flow), shish-kebabs formed in a completely quiescent (unstrained) melt with the stress self-generated by crystallization, and increased cavitation in NP-containing samples, attributed to silica restricting chain mobility. As a 'possible explanation' for the NP pushing they proposed a depletion layer (a particle-free zone created near the crystal front because polymer chains lose configurational entropy near a surface) ahead of the growth front whose high negative pressure pulls particles in; a qualitative 2D finite-element solution of the Navier–Stokes equations (in Mathematica) illustrated the flow and isobar patterns. A general review of spherulite science up to that date is Crist and Schultz (Prog. Polym. Sci. 56, 1–63, 2016).

These observations are to be distinguished from the adjacent, by-now quantitative literature on crystallization-directed assembly of small nanoparticles (tens of nm), where the criterion is simply that NP diffusion outruns the crystal growth rate, so NPs are expelled into interlamellar amorphous regions (e.g. Zhao et al., ACS Cent. Sci. 3, 751–758, 2017; Krauskopf, Jimenez, Altorbaq, Müller and Kumar, ACS Nano 15, 14430–14443, 2021, which quantified the ordered interlamellar NP state by small-angle X-ray scattering (SAXS) correlation analysis). That regime does not apply to the 200 nm particles accumulated at interspherulite boundaries in the Nat. Commun. work, precisely because their diffusion is ~1000× too slow — a different, unexplained transport mechanism must operate there. The dynamics of nanoparticles in polymer melts across length and time scales has been reviewed theoretically by Ge (Macromolecules 56, 3809–3837, 2023), but without treating moving crystallization fronts.

Problem Statement

Establish a quantitative, predictive account of crystallization-induced negative pressure and the associated melt flow in confined semicrystalline polymer films and their nanocomposites, sufficient to explain the 3D morphologies observed by two-photon optical tomography (Yang et al., Nat. Commun. 12, 5054, 2021). Concretely, the account must quantitatively address each of the following open issues: (1) the physical mechanism and quantitative force balance by which relatively large (~200 nm) silica nanoparticles are rejected by and transported ahead of growing spherulite fronts and deposited at spherulite boundaries — decomposing the relative contributions of diffusion (known to be ~1000× too slow via the Stokes–Einstein relation, which sets the diffusion coefficient of a sphere of radius rr in a melt of viscosity η\eta as D=kBT/(6πηr)D = k_BT/(6\pi\eta r)), hydrodynamic pull from the crystallization-induced negative pressure, particle–polymer interactions, and the proposed depletion layer ahead of the growth front; (2) the predictive, quantitative linkage between crystallization-induced negative pressure and melt flow (so far illustrated only by qualitative 2D finite-element Navier–Stokes calculations) and the observed consequences — mirrored spherulite nucleation across films, shish-kebab formation in quiescent (unstrained) melt, spherulite asphericity near confining surfaces, cavitation, and boundary-localized mechanical weakening — including transient pressure/flow fields during crystallization; (3) the parameter map: how molecular weight, polymer chemistry and mobility (e.g., glass transition temperature), film thickness, and nanoparticle type, concentration, size, and surface coating control the propensity for mirrored nucleation and the vertically aligned columns of spherulites that mirrored nucleation produces, the onset and spatial extent of quiescent shish-kebab nucleation, and the degree of cavitation and boundary weakening. Each of the three issues is to be treated quantitatively; for issue (3) the parameter map covers a bounded, explicitly stated subset of the listed parameters and morphologies rather than their full cross-product. The paper also flags methodological caveats concerning the fluorescence contrast itself (dye blending vs. post-crystallization infiltration, dye back-diffusion, and the influence of labeled nanoparticles on chain mobility); resolving these would strengthen any quantitative account of the native morphology, but they are presented in the paper as observational caveats rather than as posed open research questions.

References

  1. Shu-Gui Yang, Zhen-Zhen Wei, Liliana Cseh, Pantea Kazemi, Xiang-bing Zeng, Hui-Jie Xie, Hina Saba, Goran Ungar, "Bowls, vases and goblets—the microcrockery of polymer and nanocomposite morphology revealed by two-photon optical tomography", Nature Communications 12, 5054 (2021). DOI: 10.1038/s41467-021-25297-w, https://www.nature.com/articles/s41467-021-25297-w (primary source)
  2. Shu-Gui Yang, Hui-Jie Xie, Hina Saba, Liliana Cseh, Goran Ungar, "Fluorescence microscopy tracking of dyes, nanoparticles and quantum dots during growth of polymer spherulites", Polymer 191, 122246 (2020). DOI: 10.1016/j.polymer.2020.122246, https://doi.org/10.1016/j.polymer.2020.122246
  3. Mingce Tian, Wenjuan Zhou, Weijiang Guan, Chao Lu, "Real-Time Imaging of Stress in Single Spherulites and Its Relaxation at the Single-Particle Level in Semicrystalline Polymers", Analytical Chemistry 94, 17716–17724 (2022). DOI: 10.1021/acs.analchem.2c04683, https://doi.org/10.1021/acs.analchem.2c04683
  4. Alejandro A. Krauskopf, Andrew M. Jimenez, Abdullah S. Altorbaq, Alejandro J. Müller, Sanat K. Kumar, "Quantifying Nanoparticle Ordering Induced by Polymer Crystallization", ACS Nano 15, 14430–14443 (2021). DOI: 10.1021/acsnano.1c03850, https://doi.org/10.1021/acsnano.1c03850
  5. Shu-Gui Yang, Liang-Qing Zhang, Jiaming Cui, Xiang-bing Zeng, Baolin Guo, Feng Liu, Goran Ungar, "Morphology of Shear-Induced Polymer Cylindrites Revealed by 3D Optical Imaging", Macromolecules 56, 198–206 (2023). DOI: 10.1021/acs.macromol.2c01433, https://doi.org/10.1021/acs.macromol.2c01433
  6. Shu-Gui Yang, Liang-Qing Zhang, Changlong Chen, Jiaming Cui, Xiang-bing Zeng, Liying Liu, Feng Liu, Goran Ungar, "3D Morphology of Different Crystal Forms in β-Nucleated and Fiber-Sheared Polypropylene: α-Teardrops, α-Teeth, and β-Fans", Macromolecules 56, 5502–5511 (2023). DOI: 10.1021/acs.macromol.3c00788, https://doi.org/10.1021/acs.macromol.3c00788
  7. Jiaming Cui, Shu-Gui Yang, Ruibin Zhang, Yu Cao, Yubo Wang, Xiangbing Zeng, Feng Liu, Goran Ungar, "Continuous Spectrum of Morphologies and Phase Behavior across the Contact Zone from Poly(l-lactide) to Poly(d-lactide): Stereocomplex, Homocrystal, and Between", Macromolecules 56, 8754–8766 (2023). DOI: 10.1021/acs.macromol.3c01815, https://doi.org/10.1021/acs.macromol.3c01815
  8. Yucheng He, Yang Wang, Shuxian Liu, Hao Li, Zhenyang Luo, "Quantitative Tammann analysis of crystal nucleation in three-dimensional space of poly(L-lactic acid) by modeling with the Avrami theory", Polymer 364, 130676 (2026). DOI: 10.1016/j.polymer.2026.130676, https://doi.org/10.1016/j.polymer.2026.130676
  9. Ting Ge, "Scaling Perspective on Dynamics of Nanoparticles in Polymers: Length- and Time-Scale Dependent Nanoparticle–Polymer Coupling", Macromolecules 56, 3809–3837 (2023). DOI: 10.1021/acs.macromol.3c00260, https://doi.org/10.1021/acs.macromol.3c00260
  10. A. Galeski, L. Koenczoel, E. Piorkowska, E. Baer, "Acoustic emission during polymer crystallization", Nature 325, 40–41 (1987). DOI: 10.1038/325040a0, https://doi.org/10.1038/325040a0
  11. A. Pawlak, A. Galeski, "Stability of spherulite growth rate", Journal of Polymer Science Part B 28, 1813–1821 (1990). DOI: 10.1002/polb.1990.090281012, https://doi.org/10.1002/polb.1990.090281012
  12. Buckley Crist, Jerold M. Schultz, "Polymer spherulites: A critical review", Progress in Polymer Science 56, 1–63 (2016). DOI: 10.1016/j.progpolymsci.2015.11.006, https://doi.org/10.1016/j.progpolymsci.2015.11.006
  13. Dan Zhao, Vianney Gimenez-Pinto, Andrew M. Jimenez, Longxi Zhao, Jacques Jestin, Sanat K. Kumar, Brooke Kuei, Enrique D. Gomez, Aditya Shanker Prasad, Linda S. Schadler, Mohammad M. Khani, Brian C. Benicewicz, "Tunable Multiscale Nanoparticle Ordering by Polymer Crystallization", ACS Central Science 3, 751–758 (2017). DOI: 10.1021/acscentsci.7b00157, https://doi.org/10.1021/acscentsci.7b00157