Shu-Gui Yang, Xiang-bing Zeng*, Feng Liu, Goran Ungar*. Physical Review Letters 2026, 136 (1), 018101

Poly(lactic acid) (PLA) is meant to be the prime biodegradable and biosourced substitute for classic polymers such as polypropylene. But crystallization of its desirable 𝛼-form is too slow for fast commercial processing, even slowing down instead of accelerating with increasing supercooling. So products end up brittle in the inferior 𝛼’ form with low crystallinity. The crystallization slow-down is now explained by a new phenomenon called “polymorphic self-poisoning”. In 𝛼-form chains follow a strict up-down order, while in the less stable 𝛼’ orientation is random. But the polymer molecules don’t “know” that, and on cooling, as 𝛼’ is nearing stability, the randomly attached chains “think” they are already stable. So they linger at the surface of 𝛼 crystal blocking its growth, before eventually melting away. “Poisoning” is the term used for crystal growth blocked by attached impurities. But here the “impurities” are the native molecules, just in wrong orientation. While small molecules in melt change orientation easily, polymer chains do not. Suggested solution: design symmetric chain plastics. Preliminary work suggests that “polymorphic self-poisoning” also happens in other polymers, particularly at high supercooling encountered in fast commercial processing.

Figure 1. Spherulite radial growth rate vs 𝑇𝑐 for PLLA-9k, PLLA-36k, and PLLA-110k, and for PLLA-9k containing 30% plasticizer. Values for PLLA-36k and PLLA-110k were multiplied by 2 and 6; PLLA-36k and PLLA-9k datasets were shifted vertically as indicated.

Figure 2. X-ray data for the three polymers as a function of 𝑇𝑐. Top: SAXS long period. Bottom: 110/200WAXS lattice spacing. (a) PLLA-9K, (b) PLLA-36k, and (c) PLLA-110K. The data for plasticized PLLA-9k are shown as red empty circles in (a). All the samples were crystallized isothermally in DSC apparatus, then quenched at room temperature, recorded at room temperature.

Figure 3. (a)–(d) Schematic models used in simulation of measured growth rate. (a) Growth of 𝛼 phase happens by attachment of a new 𝛼 stem (black single arrow) to a clean unpoisoned growth front. The attachment rate is A and the detachment rate is B. (b) An 𝛼′ stem (red double arrow) can attach to the 𝛼-phase growth front, with attachment rate A′ and detachment rate B′. (c) Poisoning of 𝛼 growth as further 𝛼′ stems can attach to the poisoned surface, but 𝛼 stems cannot. (d) 𝛼′ phase can grow through thickening, but this is assumed to happen only when the number of 𝛼′ stems at the poisoned growth front is over a critical value m_c. (e) Simulated growth rate of 𝛼 and 𝛼′ phases and total, respectively.

First Author: Assoc. Prof. Yang Shugui, Xi’an Jiaotong University

Correspondence Authors: Prof. Xiang-bing Zeng, The University of Sheffield, Prof. Goran Ungar, Xi’an Jiaotong University

Full Text Link: https://journals.aps.org/prl/abstract/10.1103/yf56-tfhd