Stereosequenced crystalline polyhydroxyalkanoates from diastereomeric monomer mixtures

See allHide authors and affiliations

Science  08 Nov 2019:
Vol. 366, Issue 6466, pp. 754-758
DOI: 10.1126/science.aax8466

Toward less brittle degradable plastic

Bacteria produce a class of polyesters, termed polyhydroxyalkanoates, that are particularly appealing on account of how easily they undergo biodegradation. Unfortunately, these polymers also tend to be overly brittle for many applications. Tang et al. report that a molecular lanthanide catalyst can sequentially polymerize chiral and then achiral diastereomers to form polyhydroxyalkanoate varieties that are substantially more ductile. The catalyst selectively produces these copolymers directly from diastereomeric mixtures of monomers, obviating the need for a wasteful separation process ahead of time.

Science, this issue p. 754


Stereoselective polymerization of chiral or prochiral monomers is a powerful method to produce high-performance stereoregular crystalline polymeric materials. However, for monomers with two stereogenic centers, it is generally necessary to separate diastereomers before polymerization, resulting in substantial material loss and added energy cost associated with the separation and purification process. Here we report a diastereoselective polymerization methodology enabled by catalysts that directly polymerize mixtures of eight-membered diolide (8DL) monomers with varying starting ratios of chiral racemic (rac) and achiral meso diastereomers into stereosequenced crystalline polyhydroxyalkanoates with isotactic and syndiotactic stereodiblock or stereotapered block microstructures. These polymers show enhanced ductility and toughness relative to polymers of pure rac-8DL, subject to tuning by variation of the diastereomeric ratio and structure of the 8DL monomers.

The stereomicrostructure that defines the tacticity of polymers bearing stereogenic centers in their backbone largely determines their physical and mechanical properties and, thus, their practical performance (15). Catalyst-site–controlled stereoselective coordination polymerization is commonly used to produce high-performance stereoregular crystalline polymeric materials from chiral or prochiral monomers (69). Monomers carrying two stereogenic centers are typically separated beforehand into distinct chiral racemic (rac) (1016) and achiral meso (1719) diastereomers to produce crystalline isotactic (it) and syndiotactic (st) materials, respectively, potentially wasting substantial material and incurring considerable added energy cost associated with the separation and purification. In this context, eight-membered cyclic diolide (8DLMe; the superscripted Me denotes methyl substituents on the 8DL ring), a cyclic dimer of 3-hydroxybutyric acid (3HB), exists as the rac (R,R and S,S) and meso (R,S) diastereomers rac-8DLMe and meso-8DLMe (Fig. 1A). This monomer class is noteworthy, as we have shown that rac-8DLMe can be polymerized isoselectively via metal-catalyzed coordination ring-opening polymerization (ROP) by racemic (R,R and S,S) yttrium (Y) complexes 1 to 3 supported by C2-symmetric salcy ligands (Fig. 1C) to highly crystalline it-polymers with high to quantitative isotacticity ([mm] triad > 99%) (20). The rac-8DLMe–derived it-polymer has the same chemical structure as poly(3-hydroxybutyrate) (P3HB), the most prominent member of the polyhydroxyalkanoate (PHA) family.

Fig. 1 Diastereo- and enantioselective polymerization.

(A) Individual stereoselective ROP of rac- and meso-8DLMe into it- and st-P3HB. [cat], catalyst. (B) Designed diastereo- and enantioselective ROP of a diastereomeric mixture of rac/meso-8DLMe into it- and st-stereodiblock P3HB, it-sb-st-P3HB. (C) Chemical structures of the four metal-based racemic catalysts chosen for this study.

PHAs are naturally produced by bacteria and other microorganisms from biorenewable resources (2129) and are biodegradable in ambient environments, even in the ocean, thus displaying better end-of-life options relative to those for plastics that require controlled, industrial composting conditions to break down (30). Considering the generally recognized better scalability, more rapid catalyst tuning to accommodate diverse substrate structures, and faster reaction kinetics of chemical catalysis approaches relative to biosynthetic pathways, we sought to develop the catalytic chemical route to PHAs from 8DL monomers, which could potentially promote broader applications of PHAs as commodity plastics. The other chemical route to it-P3HB is through the ROP of rac-β-butyrolactone (β-BL) carrying only one stereogenic center, but it yielded P3HB with much lower isotacticity and melting transition temperature (Tm) (3137) relative to the bacterial P3HB or the P3HB produced by the ROP of rac-8DLMe with two stereogenic centers. However, although P3HB is biorenewable, biodegradable, and biocompatible and exhibits desirable physical properties (high crystallinity and Tm) and barrier properties (as excellent packaging material), it is mechanically brittle (~3% elongation at break) (38, 39), which largely limits its otherwise many potential applications. Although bacterial PHAs can incorporate various short and long alkyl side groups into copolymers with improved ductility relative to P3HB, the resulting copolymers exhibit diminished crystallinity and mechanical strength due to the randomly placed comonomer units in the typical bacterial copolymers. Hence, the chemical catalysis route could be a solution for providing PHAs with more desirable properties such as combined high crystallinity and ductility via catalyst-controlled stereoselective (co)polymerization of various 8DLR monomers to PHAs with stereosequenced microstructures.

The current synthesis of rac-8DLMe proceeds through a four-step process starting from dimethyl succinate (20, 40) to yield a mixture of the rac and meso diastereomers in an overall yield of ~38% based on the succinate or ~60% based on the first intermediate of the process, commercially available dimethyl-2,5-dioxocyclohexane-1,4-dicarboxylate (see supplementary text). Removal of meso-8DLMe by fractional crystallization or column chromatography from the diastereomeric mixture led to pure rac-8DLMe in ~40% based on the starting mixture. Thus, it would be preferable if a stereoregular, crystalline PHA could be obtained directly from the ROP of a diastereomeric 8DL monomer mixture, but such ROP would lead to a stereorandom, amorphous polymer unless the catalyst were diastereoselective. We hypothesized that, in a catalyst-site–controlled ROP, if the racemic catalyst were to mediate living and stereoselective polymerization of rac- and meso-8DLMe to it- and st-P3HB, respectively (Fig. 1A), while exhibiting both high enantioselectivity between the rac-8DLMe stereoisomers as well as high diastereoselectivity between rac-8DLMe and meso-8DLMe, then the ROP of the diastereomeric mixture could afford crystalline it- and st-stereodiblock P3HB, it-sb-st-P3HB (Fig. 1B). In such a postulated system, the (R,R)-catalyst would selectively polymerize (S,S)-8DLMe, whereas the (S,S)-catalyst would polymerize (R,R)-8DLMe. When rac-8DLMe was (almost) consumed, chain propagation would cross over to addition of meso-8DLMe to form (tapered) stereodiblock P3HB.

At the outset, we first examined stereoselective ROP reactions of preseparated rac-8DLMe and meso-8DLMe using racemic yttrium complexes 1 to 3. We previously reported that such complexes, particularly racemic-3 with bulky trityl groups at the ortho-phenoxy positions of the salcy ligand, catalyzed polymerization of rac-8DLMe to pure it-P3HB with Tm up to 175°C, number-average molecular weight (Mn) up to 1.54 × 105 g mol−1 (Da), and dispersity (Đ) as low as 1.01 (20). The resulting it-P3HB was shown to be a mixture of enantiomeric poly[(R)-3HB] and poly[(S)-3HB], owing to the complete enantioselectivity of the (R,R)-catalyst for addition of (S,S)-8DLMe and the (S,S)-catalyst for addition of (R,R)-8DLMe and no transesterification upon full monomer conversion. These prior results indicated that such racemic catalysts should also be stereoselective toward meso-8DLMe, as the (R,R)-catalyst should selectively ring-open meso-8DLMe at the (S)-site while the (S,S)-catalyst should selectively cleave the ester bond of meso-8DLMe at the (R)-site, thus affording st-P3HB (Fig. 1A), currently produced by the ROP of β-BL (31, 4144). Guided by this hypothesis, we studied the respective catalytic activities of complexes 1 to 3 [combined with 1 equivalent (equiv.) of benzyl alcohol (BnOH), which converts in situ the silylamide precatalyst to the alkoxide catalyst] toward stereoselective ROP of meso-8DLMe in dichloromethane (CH2Cl2) at room temperature. As summarized in table S2, complex 1 with 3,5-CMe3 substituents on the salcy ligand rapidly polymerized 100 or 200 equiv. meso-8DLMe to completion within 4 or 8 min, producing syndio-rich, crystalline P3HB with a Pr (syndioselectivity, defined as the probability of rac linkages between 3HB units) of 0.67 based on 13C nuclear magnetic resonance (NMR) analysis (fig. S17), a glass transition temperature (Tg) of 3.8°C, and a Tm of 99°C measured by differential scanning calorimetry (DSC) (fig. S18). Complex 2 with bulkier 3,5-CMe2Ph (Ph, phenyl) substituents had similarly high activity but higher syndioselectivity; for example, the ROP by 2 (0.5 mol %) produced st-P3HB with Pr = 0.72, Tm = 111°C, Mn = 76.5 kDa, and Đ = 1.04. With installation of even bulkier CPh3 substituents at the 3-positions of the ligand, complex 3 further enhanced the syndiotacticity to yield st-P3HB with Pr = 0.81 and Tm = 141°C, at the expense of reduced activity. To increase both the polymerization activity and syndioselectivity, lanthanum (La) complex 4, with the 3-CPh3-5-Me–substituted ligand complexed to the larger La ion, was synthesized and exhibited both good activity and high syndioselectivity, affording st-P3HB with a high Tm of 170°C and Pr ~ 0.92 (table S2).

With well-controlled isoselective polymerization of rac-8DLMe and syndioselective polymerization of meso-8DLMe firmly established and apparent differential rates observed with the C2-ligated Y and La catalysts, we set out to test the hypothesis of diastereo- and enantioselective polymerization for the possible synthesis of it- and st-stereodiblock P3HB when rac-8DLMe and meso-8DLMe were polymerized together in a one-pot fashion. We used NMR to monitor polymerization of rac/meso-8DLMe (1/1 ratio) in room temperature CH2Cl2 by 3 (1.0 mol %) and found that, as expected, rac-8DLMe was rapidly consumed first; within 30 s, rac-8DLMe had essentially been quantitatively (>97%) consumed, at which point 24% meso-8DLMe was also consumed (Fig. 2A). Extending the reaction to 35 min led to near-quantitative (>98%) conversion of meso-8DLMe, giving rise to the final polymer as tapered stereodiblock P3HB, it-P3HB-sb-st-P3HB (see Fig. 1B). This proposed stereodiblock structure was confirmed by the following six lines of corroborative evidence. First, kinetics showed a substantially higher rate of polymerization of rac-8DLMe relative to meso-8DLMe under identical conditions (Fig. 2A), with an apparent first-order rate constant ratio krac/kmeso of 82 (fig. S22). The monomer reactivity ratios measured in the copolymerization (table S5 and fig. S23) also showed a tendency for comonomers to form long blocky segments, as in the case of stereotapered diblock P3HB. Second, the isolated it-sb-st-P3HB was a semicrystalline material with a Mn of 22.2 kDa, a Tg of 1.3°C, two crystallization temperatures (Tc) at 87° and 69°C, as well as two Tm at 135° and 115°C (Fig. 2B), characteristic of a segmented block polymer containing crystalline it- and st-domain structures. The dispersity of the polymer (Đ = 1.01; fig. S66) was extremely narrow, indicative of minimal transesterification (vide infra). Third, the properties of the stereoblock polymer could be modified by changing the rac/meso ratio of the diastereomer mixture, the [8DLMe]/[3] ratio, and the catalyst (table S3). For example, using a 2/1 rac/meso ratio led to a stereoblock polymer with essentially the same Mn and Tg values but Tm values rising to 144° and 126°C, whereas keeping this 2/1 rac/meso ratio but increasing the [8DLMe]/[3] ratio from 100/1 to 200/1 further enhanced Mn to 34.6 kDa (Đ = 1.02) and Tm to 150° and 133°C (Fig. 2C). Switching the catalyst to La 4 produced a stereoblock P3HB with the highest Tm values of 157° and 141°C (fig. S21).

Fig. 2 Selected evidence for the formation of tapered it- and st-stereodiblock P3HB.

(A) Time-conversion plots in the polymerization of rac/meso-8DLMe [1/1 ratio, 1.0 M in CH2Cl2, room temperature (rt), [8DLMe]/[3] = 100/1]. (B) DSC curve of stereodiblock P3HB produced by 3 [rac/meso = 1/1, [8DLMe]/[3] = 100/1, rt, CH2Cl2]. (C) DSC curve of stereodiblock P3HB produced by 3 [rac/meso = 2/1, [8DLMe]/[3] = 200/1, rt, CH2Cl2]. (D) DSC curve of it-P3HB-sb-st-P3HB stereodiblock control prepared by sequential block copolymerization by 3 [rac/meso = 2/1, [8DLMe]/[3] = 200/1, rt, CH2Cl2].

Fourth, the structure of the semicrystalline it- and st-tapered stereodiblock P3HB produced by 3 was characterized by powder x-ray diffraction (PXRD), and crystalline diffraction patterns of the respective it- and st-P3HB domains were both evident (fig. S79). Fifth, matrix-assisted laser desorption/ionization (MALDI)–time-of-flight mass spectroscopy was used to characterize the chain and chain-end structures of low–molecular weight samples. When a low monomer/catalyst ratio of [rac-8DLMe]/[meso-8DLMe]/[3] = 10/10/1 was used and the ROP was stopped at a short reaction time of 0.5 or 1 min (at which time the conversion of rac-8DLMe was close to 100% and meso-8DLMe was 76 or 91%), the MALDI mass spectrum of the polymer nearly exclusively displayed spacings between neighboring molecular ion peaks characteristic of 8DLMe and initiation and termination chain ends characteristic of BnO and H (figs. S27 and S28). Hence, after full conversion of rac-8DLMe but before full conversion of meso-8DLMe, no transesterification side reaction had occurred, and tapered stereodiblock P3HB (it-P3HB-sb-st-P3HB) should have formed. Sixth, target stereodiblock it-P3HB-sb-st-P3HB was prepared by sequential ROP of rac-8DLMe, followed by meso-8DLMe. In a typical ROP by 3 (0.5 to 1.0 mol %), rac-8DLMe was completely consumed in <2 min to produce the it-P3HB block in a living fashion; the ensuing ROP of the added meso-8DLMe was considerably slower, requiring 20 min to 2 hours (1.0 mol % 3) to achieve full conversion (table S6). The resulting stereoblock polymer exhibited low dispersities of 1.01 to 1.02 (figs. S71 to 74) and, characteristically, two Tm values for each polymer (Fig. 2D and fig. S29). Relative to the stereotapered it-sb-st-P3HB prepared by the direct ROP of the rac/meso mixture, the two Tm values of the stereodiblock polymer by this sequential ROP procedure were noticeably higher (typically by 6° to 10°C; compare Fig. 2D with Fig. 2C); we attribute this discrepancy to the absence of the stereotapered junctions associated with incorporation of meso-8DLMe units into the it-P3HB block and vice versa. Nonetheless, the independently synthesized stereodiblock P3HB and the stereotapered it-sb-st-P3HB exhibited essentially identical PXRD patterns (fig. S80).

To examine the effects of the stereoblock microstructures of the P3HB materials on the mechanical properties, we performed tensile testing of dog-bone–shaped stereoblock polymer specimens. The resulting stress-strain curve (Fig. 3A) of the above-described stereotapered it-P3HB-sb-st-P3HB material synthesized from the ROP of 1/1 rac/meso 8DLMe with catalyst 3 in CH2Cl2 yielded ultimate tensile strength (σ) = 9.7 ± 0.5 MPa, Young’s modulus (E) = 317 ± 39 MPa, and elongation at break (ε) = 17 ± 5% (the data included the results obtained from twice-recycled specimens prepared by compression molding to gain information about reprocessability; as expected, virgin specimens produced the best mechanical performance). Thus, the formation of it-P3HB-sb-st-P3HB enhanced the ductility by about a factor of 6 relative to it-P3HB (ε = ~3%). The highly isotactic (Pm = 0.97), crystalline (Tm = 108°C) poly(3-hydroxyvalerate) (it-P3HV) prepared from the ROP of rac-8DLEt by the current catalyst system (table S7) was as brittle as it-P3HB, with σ = 9.6 MPa, E = 587 MPa, and ε = 2.3% (Fig. 3A, black curve). Decreasing the st-P3HB fraction in it-P3HB-sb-st-P3HB from 50% (by catalyst 3) to 14% (by catalyst 4) lowered the ε value from 17 to 2.9%, whereas the σ and E values increased to 16.0 MPa and 1.16 GPa, respectively (blue versus red curves in Fig. 3A), thus highlighting the importance of the st-fraction and its Pr in the mechanical performance of such materials.

Fig. 3 Mechanical properties of stereoblock P3HB polymers and copolymers.

(A) Stress-strain curves of it-P3HB-sb-st-P3HB (Mn = 22.2 kDa, Đ = 1.01, it/st ~ 1:1) synthesized from the ROP of 1:1 rac/meso-8DLMe with catalyst 3 in CH2Cl2 (blue curve); it-P3HV (Mn = 31.8 kDa, Đ = 1.18) synthesized from the ROP of rac-8DLEt with catalyst 2 in CH2Cl2 (black curve); and it-P3HB-sb-st-P3HB (Mn = 138 kDa, Đ = 1.02, it/st ~ 6:1) synthesized from the ROP of 3:1 rac/meso-8DLMe with catalyst 4 in CH2Cl2 (red curve). (B) Stress-strain curve of st-P3HB-sb-it-P3HBV (Mn = 113 kDa, Đ = 1.27, 18% 8DLEt incorporation) synthesized from the ROP of meso-8DLMe and rac-8DLEt (4/1 ratio) with catalyst 2 in THF (green curve). Extension rate = 10 mm/min, ambient conditions, break point indicated by “X.”

To further enhance the ductility, and thus the toughness, of the stereoblock P3HB materials, copolymerization of a diastereomer of 8DLMe (rac or meso) with a diastereomer of ethyl-substituted 8DL (8DLEt) was investigated, with the aim of generating a PHA as 8DLMe/Et copolymer—namely, poly(3-hydroxybutyrate-co-3-hydroxyvalerate) (P3HBV). Initial attempts by copolymerizing rac-8DLMe and meso-8DLEt with catalyst 3 in CH2Cl2 led to a semicrystalline polymer with two high Tm values of 152° and 143°C (fig. S38) corresponding to it-P3HB and it-P3HBV blocks with some incorporation of meso-8DLEt. The formation of such material is a result of largely differentiating reactivity between these two diastereomers: Kinetic studies showed that the rac-isomer and the Me-substituted monomer were far more reactive than the meso-isomer and the Et-substituted monomer (fig. S32). To overcome this double reactivity effect, we hypothesized that copolymerization of meso-8DLMe and rac-8DLEt by a selective catalyst in a suitable solvent should lead to gradient stereoblock copolymer st-P3HB-sb-it-P3HBV according to the scenario outlined in Fig. 4A. With the use of less-bulky catalyst 2, the copolymerization in CH2Cl2 led to an amorphous copolymer, owing to only slightly faster ROP of meso-8DLMe than rac-8DLEt (table S9 and fig. S39). Switching to tetrahydrofuran (THF) brought about more pronounced rate differentiation (fig. S34). Although THF has similar polarity to CH2Cl2, its coordinating ability (to the catalyst center) appears to selectively hinder coordination and/or incorporation of larger rac-8DLEt such that meso-8DLMe was more rapidly consumed first, followed by gradual, slower, but nonetheless complete consumption of rac-8DLEt (fig. S34). Accordingly, this copolymerization in THF produced a semicrystalline copolymer with Mn = 25.2 kDa (Đ = 1.23) and two broad Tm values of 112° and 94°C (fig. S40), corresponding to the syndio-rich P3HB block with some rac-8DLEt incorporation and the iso-rich P3HBV block (Fig. 4A). This composition is also consistent with the PXRD patterns (fig. S81) and the kinetics of this polymerization obtained from the time-conversion plots of the mixed diastereomers (fig. S34). To lower 8DLEt incorporation while increasing the molecular weight of the stereoblock copolymer for mechanical testing, we carried out the same copolymerization of meso-8DLMe and rac-8DLEt in a 4/1 ratio; this protocol afforded a high–molecular weight crystalline st-P3HB-sb-it-P3HBV (Mn = 113 kDa, Đ = 1.27, Tm = 135°C, Tg = 1.7°C, 18% 8DLEt incorporation; table S9). Tensile testing of this gradient stereoblock copolymer yielded impressive mechanical properties with σ = 24.1 ± 1.5 MPa, E = 169 ± 9 MPa, and ε = 564 ± 25% (Fig. 3B), thus revealing a strong, ductile, and tough material.

Fig. 4 Stereoblock copolymerization of mixed diastereomers of different 8DL monomers.

(A) Formation of stereogradient block copolymer st-P3HB-sb-it-P3HBV consisting of the syndio-rich P3HB block with some rac-8DLEt incorporation and the iso-rich P3HBV block from the copolymerization of meso-8DLMe and rac-8DLEt with a selective catalyst. (B) Formation of stereoblock copolymer it-P3HBV-sb-st-P3HBV consisting of the iso-rich P3HBV block and the syndio-rich P3HBV block from copolymerization of rac/meso-8DLMe/8DLEt with a selective catalyst.

The most convenient yet most challenging way of constructing stereosequenced block copolymers of 8DLMe/Et is through direct copolymerization of all diastereomers of 8DLMe/Et together. Figure 4B outlines a postulated scenario for the synthesis of stereoblock copolymer it-P3HBV-sb-st-P3HBV consisting of iso-rich P3HBV and syndio-rich P3HBV blocks, on the basis of the above-established reactivity trend of rac-8DLMe > rac-8DLEt, meso-8DLMe > meso-8DLEt, and the catalyst’s enantioselectivity toward the rac monomers and diastereoselectivity toward the rac/meso isomers. Under given catalyst and solvent conditions, the relative reactivity of rac-8DLEt versus meso-8DLMe can be tuned; however, if the reactivity difference becomes sufficiently small, then the junction linking the two it- and st-blocks would be largely stereorandom. To test this hypothesis, the ROP of the mixture containing all six diastereomers in an equal ratio of 8DLMe and 8DLEt was performed by catalyst 3 in CH2Cl2, yielding a semicrystalline copolymer with two Tm values on DSC curves (figs. S42 and S43). The reaction kinetics (table S10 and figs. S35 and S36) is consistent with a stereogradient diblock copolymer consisting of iso-rich P3HB and syndio-rich P3HB blocks with some 8DLEt incorporation in both blocks, thus best described as it-P3HBV-sb-st-P3HBV. Switching to more-reactive catalyst 4 afforded semicrystalline it-P3HBV-sb-st-P3HBV with Mn = 19.2 kDa (Đ = 1.07), 8DLEt incorporation = 20 mol %, and two Tm values of 125° and 110°C (table S9 and fig. S44). To further examine the scope of the 8DL substituents, the n-butyl–substituted 8DL compounds rac-8DLBu and meso-8DLBu were copolymerized with rac-8DLMe and meso-8DLMe (1/1/1/1 ratio) by catalyst 4 (0.5 mol %, 1 equiv. BnOH) in CH2Cl2 at room temperature, affording semicrystalline PHA as block copolymer it-P3HBHp-sb-st-P3HBHp [P3HBHp = poly(3-hydroxybutyrate-co-3-hydroxyheptanoate)] with Mn = 20.4 kDa (Đ = 1.10), 8DLBu incorporation = 15 mol %, and two Tm values of 127° and 105°C (table S11 and fig. S45).

The above results demonstrated that catalyst-site–controlled diastereoselective polymerization methodology enabled direct polymerization of diastereomeric mixtures of the same or different 8DLR monomers (R = Me, Et, n-Bu) into stereosequenced crystalline PHAs with tunable material properties by varying the catalyst and monomer structures and the ratio of starting rac/meso diastereomers. However, before this catalytic chemical route to PHAs can be practiced at scale, at least two remaining challenges must be addressed: (i) the development of economically more-viable preparations of the 8DLR monomers from inexpensive building blocks, ideally from the chemicals resultant from deconstruction or degradation of PHAs toward a circular life cycle, and (ii) the discovery of more-active and more-selective catalysts for meso diastereomers and 8DLR monomers with bulky R groups so that stereosequenced PHAs with higher molecular weights and crystallinity as well as better mechanical performance can be achieved.

Supplementary Materials

Materials and Methods

Supplementary Text

Figs. S1 to S81

Tables S1 to S11

References (4550)

References and Notes

Acknowledgments: This work made use of the Central Instrument Facilities of Colorado State University and the polymer characterization instruments of the Chen and G. Miyake research groups. Funding: This work was supported in part by the U.S. National Science Foundation (NSF-1664915) and the U.S. Army Research Office (W911NF1810435). Author contributions: E.Y.-X.C. conceived the project and directed research. X.T. designed and conducted experiments related to monomer and polymer synthesis, and A.H.W. performed experiments related to monomer synthesis. X.T. and E.M.W. designed and conducted experiments related to polymer characterizations. X.T. and E.Y.-X.C. wrote the initial manuscript and revised subsequent versions of the manuscript, and all authors contributed to the revised manuscript. Competing interests: E.Y.-X.C. and X.T. are inventors on U.S. patent application U.S. 2019/0211144 A1, based on U.S. provisional applications 62/616,277 and 62/671,069, submitted by Colorado State University Research Foundation, which covers the herein described crystalline polymers and copolymers from cyclic diolides. A.H.W. and E.M.W. declare no competing interests. Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the supplementary materials.
View Abstract

Stay Connected to Science

Navigate This Article