**1. Introduction**

In recent decades, bio-based polymers derived from annually renewable resources have drawn increasing attention [1], as biomass is the only source of available renewable carbon. Among them, polylactide (PLA) is the most promising polymer for the replacement of conventional thermoplastics, especially because it is also biodegradable (compostable) and can be used for a wide range of applications, including biomedical products, textiles, daily appliances, packaging, items used in agriculture and engineering [1–5].

Amorphous PLA with the glass transition temperature, Tg, in the range of 55–60 ◦C, is stiff and brittle at room temperature. The crystallizability of PLA strongly depends on its enantiomeric composition and worsens with increasing content of repeating units of different chirality in the chain [6,7]. However, even optically pure poly(L-lactide) (PLLA), if cooled sufficiently fast, remains amorphous and vitrifies. Then, upon heating from the glassy state, such amorphous PLLA can cold-crystallize. However, not only the enantiomeric composition but also other factors, including molar mass and macromolecular architecture, for instance, branching or star structure, are also important and influence the crystallization of PLA. Depending on crystallization temperature, PLA chains crystallize from melt in the ordered alpha or disordered alpha' orthorhombic forms [8], which can be identified not only by wide angle X-ray scattering but also by Raman spectroscopy or nuclear magnetic resonance spectroscopy [9,10].

Star polymers are especially interesting for many applications, for example in biomedicine or engineering [11,12]. Their rheological, thermal and mechanical properties can differ from those of their linear counterparts [11,13]. In addition, processing of star polymers can

**Citation:** Bojda, J.; Piorkowska, E.; Lapienis, G.; Michalski, A. Shear-Induced Crystallization of Star and Linear Poly(L-lactide)s. *Molecules* **2021**, *26*, 6601. https://doi.org/ 10.3390/molecules26216601

Academic Editor: Ivan Gitsov

Received: 21 September 2021 Accepted: 22 October 2021 Published: 31 October 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

be carried out at lower temperatures than their linear analogs, which could be beneficial, especially in the case of polymers prone to thermal degradation like PLAs.

Star PLLAs were most often obtained by the bulk polymerization of cyclic lactide conducted at a temperature above the melting point of the monomer ( ∼99 ◦C), in the presence of initiators with hydroxyl end groups, with stannous octoate (Sn(Oct)2) as a catalyst [14–19]. Crystallization of star PLLAs was found to be affected by their molar masses and numbers of arms. Usually, PLLAs with number average molar masses, Mn, below 100 kg/mol were analyzed. The consequences of the relatively small molar mass of star PLLA are short arm length resulting in a small number or even absence of entanglements and a large number of chain ends as well as branching points. The presence of branching points, initiator moieties, especially bulky ones, in the middle of macromolecules, hydroxyl chain end groups enhancing hydrogen bonding, and chain directional change at the branching points, are the factors disturbing the segmental mobility [11,20]. Moreover, the branching points have to be excluded from the crystalline regions [21]. For example, for PLLAs with Mn about 35 kg/mol, nonisothermal crystallization peak temperature, Tc, and overall isothermal crystallization rate decreased upon the increasing number of arms up to six [22]. Moreover, an increase of Tg, cold crystallization peak temperature, Tcc, accompanied by a decrease of melting peak temperature, T m, and crystal growth rate of 3-arm PLLAs, with Mn in the range of 13 to 63 kg/mol, were observed, as compared with those of linear PLLAs [21]. It is worth noting that crystallization of 6-arm PLLAs was rarely examined, and the studies were limited to polymers with Mn per one arm below 10 kg/mol. For instance, it was found [23] that for 1-, 2-, 4- and 6-arm PLLAs, T m, Tcc, and crystallinity, χc, decreased with increasing number of arms at a fixed Mn. Recently, Bojda et al. [24] synthesized three star PLLAs, two 6-arm with weight average molar masses, M w, of 120 and 245 kg/mol and one 4-arm with M w of 123 kg/mol, and compared their crystallization with that of linear ones with M w of 121, 291 and 339 kg/mol. At M w close to 120 kg/mol the star architecture decreased the crystal growth rate in the temperature range of 120–145 ◦C. Crystal growth of PLLAs with M w > 200 kg/mol was the slowest and unaffected by the macromolecular architecture. The slow crystal growth in these PLLAs resulted in their weak crystallization during cooling.

In turn, it is long known that the crystallization of polymers, including PLAs, is strongly influenced by flow, which plays a vital role during industrial processing. The flowinduced macromolecular orientation can strongly affect the crystallization kinetics and the resulting structure, which are controlled by the interplay between crystallization and chain relaxation. The fundamental processes governing the flow-induced crystallization of polymer melts were discussed by many authors [25–27]. The shear-induced crystallization strongly depends on the temperature of shearing, Ts, shear rate, . *γ*, and total strain. It is believed [28] that to induce the point-like nuclei and fibrillar nuclei, the shear rate has to exceed the inverse reptation time and the inverse Rouse relaxation time of the high molar mass tail, respectively, although when the flow is strong enough, but too short, intermediate regimes were also defined. It is worth noting that others postulated that mechanical work is a controlling parameter [26,29,30]. A very important factor is the polymer molecular characteristic, especially the high molar mass tail is crucial due to the vital role of the longest macromolecules in the flow-induced crystallization. [26,31,32]. Similarly to other polymers, the enhancement of point-like nucleation, formation of oriented nuclei and shish-kebab structures induced by shear were observed in PLAs [33–35]. The effect of shear on both isothermal [34–37] and nonisothermal crystallization of PLAs was studied. Bojda et al. [36] demonstrated that smaller content of D-lactide enhanced the effect of shear on nonisothermal crystallization on PLA and that higher crystallinity degree developed during slower post-shearing cooling. In turn, Kim et al. [38] compared shear-induced crystallization of linear and 4-arm PLLAs with M w of about 2 kg/mol and found that that of the latter was slightly faster.

To the best of our knowledge, the effect of shear flow on the crystallization of star PLLAs with higher molar masses was not investigated. Only shear-induced isothermal crystallization of PLAs with long chain branching (LCB), prepared by γ irradiation, was studied in [39,40]. It was demonstrated that the shear-induced nucleation density in LCB PLA was strongly enhanced in comparison to linear PLA and increased with increasing LCB degree. Moreover, the transformation from spherulitic to oriented crystalline morphologies was observed. It was concluded that the shear-induced formation of the oriented crystalline morphology of LCB PLAs was related to the hindering of relaxation of the stretched LCB macromolecular chain network.

In the present study, shear-induced nonisothermal crystallization of star PLLAs with M w close to 120 and 240 kg/mol was studied and compared with that of their linear analogs. In addition, the crystallization of linear PLLA with M w of 339 kg/mol was also examined. The polymers were sheared at 170 and 150 ◦C and cooled at 10 or 30 ◦C/min. The crystallization was followed by a light depolarization technique, whereas the crystallized specimens were ex-situ examined with scanning electron microscopy (SEM), differential scanning calorimetry (DSC), small- and wide-angle X-ray scattering (2D-SAXS and 2D-WAXS).

## **2. Materials and Methods**

Linear PLLAs having at one chain end benzyl alcohol and star-shaped PLLAs having as a core: di(trimethylolpropane) (4-arm PLLA-OH) or dipentaerythritol (6-arm PLA-OH), were synthesized in bulk at 130 ◦C by coordination polymerization using hydroxyl compound as an initiator and stannous octoate as a catalyst, as described previously [24]. The resulting PLLAs were dissolved in dichloromethane and precipitated into methanol, separated by filtration, and washed several times with methanol. The polymers were characterized with size exclusion chromatography (SEC) and 1H NMR. After purification, the polymers were stabilized with 0.2 wt.% of Irganox 1010 and 0.2 wt.% of Irganox 1024, both from BASF (Ludwigshafen, Germany). The details of polymerization, purification, characterization and stabilization were previously described [19,24,41,42].

The average molar masses, number, Mn, weight, M w, and z-average, Mz, and dispersity of linear and star-shaped PLLAs measured by SEC in dichloromethane are given in Table 1.

**Sample Code M w (kg/mol) Mn (kg/mol) Mz (kg/mol) M w/Mn** L121 121 81 194 1.5 L240 240 157 414 1.3 L339 339 257 495 1.3 4S123 123 97 152 1.3 6S120 120 80 162 1.5 6S2452451832941.3

**Table 1.** The average molar masses, number, Mn, weight, M w, and z-average, Mz, and dispersity Mw/ Mn of star and linear PLLAs.

For studies of crystallization, 200 μm thick films were compression moulded at 200 ◦C for 3 min in a hydraulic press and quenched to room temperature, RT, between metal blocks.

Flow-induced crystallization was carried out in the Linkam CSS-450 optical shearing system (Linkam, Waterfield, UK) mounted in a polarizing light microscope (PLM) Nikon Eclipse 80i equipped with Nikon DS Fi1 video camera. The films were: heated to 210 ◦C at 30 ◦C/min and held at 210 ◦C for 3 min to erase the thermal history. Next the films were cooled at 30 ◦C/min to Ts of 170 or 150 ◦C and sheared at a rate, . *γ*, of 5, 10 and 20/s for 20, 10 and 5 s, respectively to reach the same strain of 100. After shearing, they were cooled to RT at a rate, v, of 10 or 30 ◦C/min. The shearing conditions were selected based

on preliminary studies. Control specimens were subjected to a similar thermal treatment in quiescent conditions; they were held at Ts for 20 s without shearing.

The conversion of melt into the crystalline phase was followed using the light depolarization method. The intensity of transmitted depolarized light was measured during cooling and the relative volume conversion degree, αvr(T), was calculated according to the expression:

$$
\alpha\_{\rm VT}(\rm T) = [\rm I(T) - \rm I(T\_{\rm O})] / [\rm I(T\_{\rm e}) - \rm I(T\_{\rm o})] \tag{1}
$$

where: I(T) denotes the intensity of transmitted depolarized light at temperature T, whereas To and Te are the initial temperature and the final temperature of the measurement.

In the plate-plate geometry, shear rate varies along a radius, hence for ex-situ examination, specimens were cut from the films at proper distances from the centers, at which shear rates were equal to the selected values.

Crystallinity and thermal properties of the specimens were analyzed with differential scanning calorimetry (DSC) using TA Instrument DSC TA Q20 (New Castle, USA) during heating at 5 ◦C/min from RT, under nitrogen flow.

Crystal orientation in the films was examined with 2D-WAXS in the transmission mode, using a WAXS camera coupled to an X-ray generator (sealed-tube, fine point CuK α source, Ni filtered, operating at 50 kV and 35 mA) from Philips (Eindhoven, Netherlands). The incident beam was normal to the film plane. The lamellar structure was probed with 2D small angle X-ray scattering (2D-SAXS). Kiessig-type camera with the sample-detector distance of 1.2 m was coupled to GeniX Cu-LD X-ray system from Xenocs (Grenoble, France), with CuK α source operating at 50 kV and 1 mA. The incident beam was normal to the film plane. The scattering patterns were recorded with Pilatus 100K solid-state detector from Dectris (Baden, Switzerland).

To reveal their internal structure, selected sheared PLLA specimens were cut across their thickness parallel to the shearing direction, and the exposed surfaces were analyzed with scanning electron microscopy (SEM) using Jeol JSM-5500 LV (Tokyo, Japan). Before the examination, the specimens were etched according to the known method [36,43], at 37 ◦C, in a solution of 61 mg of Trizma base, 2 mg of sodium azide and 4 mg of Proteinase K (all from Sigma-Aldrich, St. Louis, MO, USA) in 5 mL of distilled water. After appropriate washing and drying, the specimens were sputtered with gold.

## **3. Results and Discussion**
