A Strong and Rigid Coordination Adaptable Network that Can Be Reprocessed and Recycled at Mild Conditions

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLE7 Dec 2022A Strong and Rigid Coordination Adaptable Network that Can Be Reprocessed and Recycled at Mild Conditions Wen Li†, Hong-Qin Wang†, Wen-Tong Gao†, Zhangxia Wang, Pan Xu, Haibo Ma and Cheng-Hui Li Wen Li† State Key Laboratory of Coordination Chemistry, School of Chemistry and Chemical Engineering, Collaborative Innovation Center of Advanced Microstructures, Nanjing University, Nanjing 210023 †W. Li, H.-Q. Wang, and W.-T. Gao contributed equally to this work.Google Scholar More articles by this author , Hong-Qin Wang† State Key Laboratory of Coordination Chemistry, School of Chemistry and Chemical Engineering, Collaborative Innovation Center of Advanced Microstructures, Nanjing University, Nanjing 210023 †W. Li, H.-Q. Wang, and W.-T. Gao contributed equally to this work.Google Scholar More articles by this author , Wen-Tong Gao† State Key Laboratory of Coordination Chemistry, School of Chemistry and Chemical Engineering, Collaborative Innovation Center of Advanced Microstructures, Nanjing University, Nanjing 210023 School of Materials Science and Engineering, Nanjing Institute of Technology, Nanjing 211167 †W. Li, H.-Q. Wang, and W.-T. Gao contributed equally to this work.Google Scholar More articles by this author , Zhangxia Wang Jiangsu Key Laboratory of Vehicle Emissions Control, School of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210023 Google Scholar More articles by this author , Pan Xu State Key Laboratory of Coordination Chemistry, School of Chemistry and Chemical Engineering, Collaborative Innovation Center of Advanced Microstructures, Nanjing University, Nanjing 210023 Google Scholar More articles by this author , Haibo Ma Jiangsu Key Laboratory of Vehicle Emissions Control, School of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210023 Google Scholar More articles by this author and Cheng-Hui Li *Corresponding author: E-mail Address: [email protected] State Key Laboratory of Coordination Chemistry, School of Chemistry and Chemical Engineering, Collaborative Innovation Center of Advanced Microstructures, Nanjing University, Nanjing 210023 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.022.202101672 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail The investigation of covalent adaptable networks (CANs) is expanding rapidly due to the growing demand for sustainable materials, as CANs show thermoset-like behavior and yet can be reprocessed, recycled, and healed. However, most of the CANs reported so far have a trade-off between mechanical strength and reversible properties and often show performance reduction after reprocessing and/or recycling. Herein, we designed and synthesized a coordination adaptable network (CoAN) by crosslinking low-molecular-weight monomers with abundant coordination bonds. Owning to its excellent variable-stiffness property, leading to high stiffness at ambient conditions and low viscosity at elevated temperature, the as-prepared CoAN showed high mechanical rigidity but could be reprocessed rapidly and recycled at mild conditions. After reprocessing or recycling, the mechanical properties of the samples showed no performance reduction, compared with a pristine sample. Density functional theory calculations showed that free thiol ligands played a key role in reducing the activation energy for bond exchange. When used as binders for composites, the embedded carbon fibers could be recycled rapidly and still maintain the original microstructure. The material also showed temperature-sensitive dielectric and conductive properties due to the release of metal ions upon heating. Overall, such performances are superior among the CANs reported previously. Download figure Download PowerPoint Introduction Polymers are traditionally classified into thermosets and thermoplastics. Thermosets are crosslinked polymer networks with outstanding mechanical strength and solvent resistance used in durable goods, adhesives, and composites. Due to the irreversible nature of the crosslinking, thermosets need to be synthesized or cured in their desired final shape, giving a rigid material unable to reprocess or recycle. On the contrary, thermoplastics are built up from high-molecular-weight polymer chains through noncovalent interactions. As a result, thermoplastic polymers can be remolded readily, reprocessed, and recycle by heating, but show limited durability, poor resistance to deformation (creep), weathering, or dissolution. Given the growing global demands and societal incentives toward more sustainable materials, interests in developing polymers or their fiber composites that combine thermoset rigidity and thermoplastic processability and recyclability have grown exponentially. Recently, an attractive chemical strategy to introduce plasticity and recyclability in thermosets is offered by introducing exchangeable chemical bonds into polymer networks.1–7 Such polymers, denoted as covalent adaptable networks (CANs), also known as vitrimers (refer, in particular, to associative CANs),8 dynamic covalent networks (DCvNs),9 dynamic covalent polymer networks (DCPN),10 malleable thermosets11 or other names, could be reprocessed like thermoplastics due to their bond exchange capabilities. Meanwhile, they have three-dimensional network structures and could retain high performances characteristic of thermosets. To date, a variety of dynamic covalent chemistries have been implemented into thermosets such as transesterification,2,12,13 Diels–Alder chemistry,14,15 disulfide exchange,16,17 boric ester transesterification,18–21 imine exchange,22,23 and so on. While the study of CANs has matured remarkably in less than two decades, two critical issues still need to be addressed before these materials can be applied in more ambitious directions. One is the trade-off problem between mechanical strength and reversible properties. Reversible covalent bonds usually have lower bond energy than irreversible bonds, which prevent CANs from achieving satisfactory mechanical properties. By increasing the crosslinking density, the mechanical strength of CANs can be enhanced as expected.24 However, the polymer chain segments are difficult to move in these rigid systems, even at elevated temperatures, resulting in reprocessing difficulties (time-consuming and energy-intensive). Alternatively, CANs can be strengthened by incorporating permanent crosslinks or inorganic/organic fillers.25–27 However, the presence of irreversible bonds in the permanent crosslinks and between the CAN matrix and fillers limit complete chemical recycling of the material, especially when the amounts of permanent crosslinks and/or fillers exceed a critical value, whereby a percolating network is formed. The other issue is the performance reduction after reprocessing and/or recycling. As catalysts are commonly needed to enable bond exchange in otherwise inert linkages, the exogenous catalysts might degrade or leach over time, leading to the loss of reversibility of CANs. Moreover, reprocessing of CANs at high temperatures often results in side reactions due to the reactive chain-end radical promoted by heating, leading to performance reduction of the final product. Therefore, future chemistry strategies that allow maintenance or even enhancement of polymer performances after reprocessing or recycling are highly desirable. We aimed to solve these issues through coordination chemistry. Coordination bonds are natively reversible as they belong to noncovalent interactions, with the strength falling between typical covalent bonds and hydrogen bonds. The energies of coordination bonds can be readily tuned by careful selection of the combination of ligand and metal ion. Previous studies by us and others have shown that low-molecular-weight polymers, when crosslinked by weak but dense hydrogen bonds, yield mechanically robust materials.28,29 If the hydrogen bonds are substituted by stronger coordination bonds, polymer networks with higher mechanical strength are obtained. On the other hand, the reactions between metal ions and ligands are generally spontaneous; thus, the addition of catalysts could be avoided, which is helpful to suppress side reactions during reprocessing and recycling. Therefore, we envisage that crosslinking low-molecular-weight monomers with abundant coordination bonds would address the present challenges of CANs. Herein, we report a CAN based on the dynamic coordination bonds between Zinc(II) ions and pentaerythritol tetrakis(3-mercaptopropionate) (PETMP) ligands. Owning to its excellent variable-stiffness property, which leads to high stiffness at ambient condition but low viscosity at elevated temperature, the as-prepared coordination adaptable network (CoAN) showed high transparency (>85% for visible light with 1 mm thickness), high mechanical rigidity (Young’s modulus of 1.6 GPa and elongation <5% before fracturing), yet, it could be reprocessed rapidly (at 150 °C, 3 MPa for 10 min) and recycled at mild conditions (dissolve in PETMP at 60 °C for 24 h and Methyl 3-mercaptopropionate (MMP) at 25 °C for 4 h). After reprocessing or recycling, the mechanical properties of samples showed no performance reduction, compared to the pristine sample. Density functional theory (DFT) calculations showed that free thiol ligands play a key role in reducing the activation energy for bond exchange. When used as binders to prepare carbon fibers composites, the carbon fibers embedded were recycled rapidly, which still exhibited high quality. The material also showed temperature-sensitive dielectric and conductive properties due to the release of metal ions upon heating. Altogether, such performances are superior among the CANs reported previously. Experimental Methods Materials and equipment used in measurements PETMP, ZnCl2, and triethylamine (TEA) were purchased from Energy Chemical Inc. (Shanghai, China). Carbon fiber powder (CFP) and 2 mm carbon fiber were purchased from Carbon Technology Co., Ltd. (Shenzhen, China). Other chemicals and solvents were purchased from Sigma-Aldrich (St. Louis, MO, United States). All chemicals were used as received without any purification. Various-temperature 1H NMR spectra were recorded on a Bruker DRX 400 NMR spectrometer spectrometer (Bruker Corp., Billerica, MA, United States) in deuterated solvents. Thermogravimetric analysis (TGA) was performed on a simultaneous SDT 2960 thermal analyzer (TA Instrument Inc., Wood Dale, IL, United States) from 30 to 600 °C at a heating rate of 10 °C/min under a N2 atmosphere. Differential scanning calorimetry (DSC) experiments were performed using a Mettler-Toledo DSC1 STARe differential scanning calorimeter (Mettler-Toledo, Oakland, CA, United States) under a dry nitrogen atmosphere (50 mL/min). The temperature range was 10–140 °C, with a heating/cooling rate of 10 °C/min. The mechanical properties were performed using an Instron 3343 instrument (Instron Inc., Norwood, MA, United States). The scanning electron microscopy (SEM) images were obtained on an S-4800 SEM (Hitachi, Japan) at 10 kV. Fourier transform infrared (FT-IR) spectra were recorded with a Horiba Jobin-Yvon Fluorolog-3 fluorometer (Horiba Instrument Inc., West Chicago, IL, United States). Powder X-ray diffraction (PXRD) was performed on a D8 advance X-ray powder diffraction instrument (Bruker Corp., Billerica, MA, United States). Dynamic mechanical analysis (DMA) measurements were made on a dynamic mechanical analyzer (TA Instruments Q800; TA Instrument Inc., Wood Dale, IL, United States) over temperatures ranging from −50 to 150 °C. The rheological behaviors were evaluated on a TA Instruments DHR-2 system (TA Instrument Inc., Wood Dale, IL, United States). Temperature sweeps were performed with 8 mm parallel plates on circular samples 8 mm in diameter. Temperature sweeps were run from 20 to 150 °C (or from 150 to 20 °C in the cyclic test mode) at a rate of 2 °C min−1 and a frequency of 1 Hz, and the strain was set at 0.1%. Frequency sweeps were run from 0.01 to 600 rad/s, and the strain was 0.1%. Contact with the sample was maintained by the auto-compression feature set to 0.2 ± 0.15 N. The thermal conductivities of the composites were measured using a DECCA-III thermal conductivity tester (Shenzhen Decca Precision Instrument Co., Ltd., Shenzhen, China). The electrical conductivity was measured using a QE-90E high-voltage insulation resistance tester (Shanghai Tai'ou Electronics, Co., Ltd., Shanghai, China). The transmittance of the polymer was tested using a UV-2700 spectrometer (Shimadzu Europe, Duisburg, Germany), ranging from 280 to 800 nm. The refraction index of the polymer was tested on Model 2010/M prism coupler (Metricon Co., Ltd., Melbourne, Australia). Zn concentrations were measured using an inductively coupled plasma optical emission spectrometry (ICP-OES, Optima 5300; Perkin-Elmer SCIEX, Wellesley, MA, United States). The capacitance data was obtained by using Novocontrol’s Concept 80 Broadband Dielectric Spectrometer (Novocontrol Technologies GmbH & Co. KG, Montabaur, Germany). Synthesis of PETMPx-y-Zn polymer The PETMP80-20-Zn polymer was synthesized by the coordination polymerization between PETMP and Zn2+, as follows: 50.00 g (0.1023 mol) of PETMP and 8.283 g (0.0819 mol) of TEA were added to a 500 mL flask equipped with a magnetic stirrer, followed by adding 100 mL dichloromethane (DCM) to the mixture and stirring at room temperature for 3 h to ensure complete reaction of PETMP and TEA. Then 13.946 g of zinc chloride (0.1023 mol) was well-dispersed in 15 mL methanol by ultrasound. Next, the methanol solution was carefully added to the aforementioned mixture drop by drop, and the mixture was stirred overnight, followed by the addition of 50 mL DCM. After 3 min of stirring, another 50 mL DCM was added to precipitate the PETMP80-20-Zn particles. Stirring was speeded up to prevent aggregation of the particles. The mixture was then filtered using a sand core funnel. The solid product was washed three times with 20 mL methanol and then washed twice with 20 mL DCM. The chunk solid obtained on the filter paper was ground into a powdery form after the filtering and washing process. The white powder was then dried under an infrared lamp (35.60 g, yield 62.84%), which was then hot-pressed into a bulk material at 150 °C, 3 MPa for 10 min. By varying the amount of TEA [9.940 g (0.0982 mol) and 41.415 g (0.4093 mol)] and following the same synthetic process, PETMP76-24-Zn and PETMP0-100-Zn powder were, respectively, obtained. Synthesis of PETMP80-20-Zn/CFP and PETMP80-20-Zn/CF composites CFP or carbon fiber was mechanically mixed with PETMP80-20-Zn powder. Grinding and ultrasonic treatment were performed to ensure uniform mixing of CFP. The resulting mixture was hot-pressed at 150 °C, 3 MPa for 1 h to obtain PETMP80-20-Zn/CFP or PETMP80-20-Zn/CF composites. Mechanical and self-healing tests Mechanical flexural strain–stress and mechanical tensile tests were performed using an Instron 3343 instrument (Instron Inc., Norwood, MA, United States) at a rate of 10 mm/min. For all the tests, samples sizes of 60 mm length, 5 mm width, and 2 mm height were used. Three samples were tested under the three-point flexural mode for each reprocessing process. For the self-healing tests, the film was cut to make a small crack using a thin blade and placed in a 130 °C environment for 1 h. Reprocessing of PETMP80-20-Zn The PETMP80-20-Zn bulk material was ground into small pieces and dried under an infrared lamp. Then the pieces were hot-pressed at 150 °C, 3 MPa for 10 min to obtain a reprocessed sample. Leaching of PETMP80-20-Zn 0.2 g PETMP80-20-Zn powder (particle size <150 μm) and 20 mL methyl mercaptopropionate methanol solution (1.2, 1.5, or 2.0 M) were added into a 100 mL flask equipped with a magnetic stirrer. The conditions for the leaching process were set at 40 °C temperature, with a stirring speed of 200 rad/s. At 10, 20, and 40 min, 0.1 mL solution was extracted and filtered using 0.22 μm nylon syringe filters. Each filtrate was used for ICP analysis to determine the zinc concentration in the solution. The leaching efficiency was defined as the amount of zinc in the whole leaching solution over the total amount of zinc in 0.2 g PETMP80-20-Zn powder. Recycling of PETMP80-20-Zn By adding 4.00 g PETMP and 8.00 mL tetrahydrofuran (THF) to the 1 g PETMP80-20-Zn powder or pieces and mixing them thoroughly at 60 °C for 24 h, the powder dissolved in the PETMP solution. Then 0.663 g TEA and 1.116 g zinc chloride were added to recover the same raw materials. The mixture was stirred overnight, then 10 mL DCM was added. After 3 min of stirring, another 10 mL DCM was added to precipitate the PETMP80-20-Zn solid. Stirring was speeded up to prevent powder aggregation, and then the mixture was filtered using a sand core funnel. Next, the solid product was washed three times with 5 mL methanol, then twice with 5 mL DCM. The powder was consistently ground during the filtering and washing process, dried, and ground under an infrared lamp, obtaining a 3.88 g yield (70.14%). Recycling of PETMP80-20-Zn/CFP composite The PETMP80-20-Zn/CFP composite was ground into pieces. Then 4.00 g PETMP and 8 mL THF were added to 1 g PETMP80-20-Zn/CFP composite pieces to dissolve the PETMP80-20-Zn basement at 60 °C for 24 h. Black CFP and clear solution were obtained after filtering the mixture. The black CFP was washed with 5 mL DCM to remove residual PETMP and dried under an infrared lamp. Next, 0.663 g TEA and 1.116 g zinc chloride were added to the clear solution to recover the raw PETMP80-20-Zn materials. The mixture was stirred overnight. Then 10 mL DCM was added. After 3 min of stirring, another 10 mL DCM was added to precipitate the PETMP80-20-Zn powder. Stirring was speeded up to prevent the aggregation of powder. Next, the mixture was filtered using a sand core funnel. The solid product was washed three times with 5 mL methanol, then twice with 5 mL DCM. The powder was consistently ground during the filtering and washing process, dried, and ground under an infrared lamp. Recycling of PETMP80-20-Zn/CF composite A 2 mL MMP solution and 2 mL methanol were mixed with 0.2 g PETMP80-20-Zn/CF composite at room temperature for 4 h. Then the mixture was filtered and washed three times with 5 mL methanol. DFT calculations Quantum chemical calculations based on DFT were performed using Gaussian 16 Rev. A.03.30 [Zn(SMe)n]2-n was considered a simplified calculation model, whereby the zinc cation was coordinated by several thiol groups. Geometries of reactants/transition states/products were optimized with no constraints at the M06-2X/def2-TZVP level30–32 to study the dissociative and associative mechanisms. Calculations were performed in the gas phase with tight optimization, and the superfine integration grid was employed. Vibrational frequency calculations were performed to ensure the nature of the stationary points, and thermochemical values were computed at room temperature of 298.15 K using uncorrected frequencies. Special attention was paid to identifying the transition states: Initial geometries were obtained by performing potential energy surface (PES) scans, then optimizations were conducted.33 The intrinsic reaction coordinate paths34 connecting the reactants and products were calculated to ensure the transition states were correct. The Multiwfn 3.8 program35 was used to calculate Hirshfeld charge36 using the checkpoint file from the above Gaussian calculations as the input file. We also tested three other functionals (B3LYP, ωB97XD, and PBE0) for the dissociative reactions, since both B3LYP and PBE0 had been employed in combination with Grimme’s pairwise dispersion correction (D3), as well as Becke–Johnson (BJ) damping,37 as shown in Supporting Information Table S4. The results showed that the free energy barriers of the ligand-mediated dissociative reaction calculated by the four functionals were lower than those of the direct dissociative reaction. Dielectric properties test To fabricate the capacitor for the dielectric properties test, a PETMP80-20-Zn polymer film (852 μm thick) was initially hot-pressed between two pieces of polyimide film (PI film). Then the polymer film was peeled off from the PI films and cut into a wafer with a dimension of 1.98 × 1.98 cm. Two copper sheets were used as the upper and lower electrodes, respectively. The capacitor was hot-pressed again to ensure complete contact between the PETMP80-20-Zn film and the electrodes before measurement. We obtained the dielectric properties of the sample by collecting the capacitance data using Novocontrol’s Concept 80 Broadband Dielectric Spectrometer (Novocontrol Technologies GmbH & Co. KG, Montabaur, Germany) at frequencies ranging from 1 to 106 Hz and with the temperature from 25 to 150 °C. The dielectric constant was calculated from the capacitance. Results and Discussion Materials design and characterization Zinc(II) thiolate coordination chemistry has been well documented in the literature as Zn2+ was found to bind cysteine in many proteins, where it plays either a structural role to stabilize the folded protein or a catalytic role in forming an active enzyme.38–40 While protein-bound Zn2+ is normally four-coordinated and tetrahedral, the thiol ligand in cysteine introduces different binding modes with Zn2+, including ligand bridged in multinuclear sites and variable coordination geometry, resulting in a considerable number of functions with a limited set of ligands.41 Various spectroscopic characterizations have proven the coordination dynamics of zinc in proteins.41 DFT calculations have also shown that the Zn–S bonds in zinc-binding proteins are both thermodynamically stable and kinetically labile.42 Inspired by the dynamic nature of Zn(II)–S interactions in metallothionein, we selected Zn(II)–S bonds for our design. According to studies on model ligand p-Toluenethiol, various coordination geometries of Zn2+ complexes could be obtained as evidenced by the mass spectra ( Supporting Information Figures S1a–S1c). The formation of [Zn2(SPhCH3)5]nn− complexes indicated at least one ligand bridging two metal centers, corresponding with the findings in zinc-cysteine interactions in proteins. The flexibility of coordination number and geometry, likely to involve an entropic contribution, could facilitate ligand exchange reactions and reduce crystallization. Mass spectroscopic study on model complexes also evidenced the ligand exchange process ( Supporting Information Figures S1d–S1f): After adding 10 equiv p-Methoxybenzenethiol into 1 equiv zinc thiolates and mixing thoroughly at room temperature, whatever structures the original zinc thiolates possessed, the mass spectra showed a series of Zn2+ complexes derived from the ligand exchange between p-Toluenethiol and p-Methoxybenzenethiol. Based on these results, we believe that the dynamic Zn(II)-S bonds could be used to construct CoANs with excellent performances. However, based on previous reports,43,44 we envisaged that incorporation of inorganic metal-ligand coordination bonds might result in crystallization, which often produces ordered structures with poor processability, undermining its practical application. Therefore, we selected PETMP for our design as the flexibility of PETMP can reduce crystallization. By reacting PETMP with zinc chloride in the presence of TEA as the base, we obtained a coordination polymer containing Zn(II)-S bonds. Surprisingly, when 4 equiv TEA was reacted with 1 equiv PETMP, during which all the thiol groups were deprotonated by TEA, we obtained a powder sample that could not be hot-pressed even when heated to 180 °C, suggesting that the Zn(II)-thiolate bonds inside the polymer network were not active enough under this temperature. However, when we added less TEA (i.e., only deprotonate partial of the thiol groups), we obtained malleable powder samples: The less the added TEA, the lowered processing temperature, which resulted in a softer as-prepared polymer. At an extreme condition, where no TEA was added, the resulting material was a viscous fluid without any mechanical strength ( Supporting Information Figure S2). In the following discussion, the samples with different degrees of deprotonation in PETMP were labeled as PETMPx-y-Zn (x indicates that x% of the thiol groups remained protonated while y indicates y% of the thiol groups were deprotonated). By considering both the mechanical strength and the processing condition of the final materials, we selected adding 0.8 equiv TEA to prepare our material (denoted as PETMP80-20-Zn). After the reaction, the PETMP80-20-Zn network was obtained as a solid material, which could be ground into powder and hot-pressed into a transparent bulk material at 150 °C, 3 MPa within 10 min (Figure 1a). By inductively coupled plasma emission spectrometry (ICP) analysis, the zinc concentration of the material was determined to be 14.68 ± 0.70 wt % (n = 3), slightly higher than the value for metal to thiol ratio of 1:4 (11.74 wt %), indicating the possibility of bridging binding mode. The band around 2568 cm−1 in the normalized FT-IR spectra showed the presence of the thiol group before and after the reaction ( Supporting Information Figure S3).45 It is worth noting that the observed aliphatic C–H stretching vibrations at 2800–3000 cm−1 were shifted to higher values, as observed when the molecule was coordinated to a Lewis acid center.46,47 No signal from 500–550 cm−1 was observed in the Raman spectra ( Supporting Information Figure S4), indicating that no disulfide bond was formed during this reaction.48,49 No apparent aggregation of the Zn(II)-S complexes were observed in the small-angle X-ray scattering (SAXS) analysis ( Supporting Information Figure S5) and energy-dispersive X-ray spectroscopy ( Supporting Information Figure S6). The SEM results of the cross-section of the material evidenced a porous microstructure ( Supporting Information Figure S6). The PXRD pattern indicates poor crystallinity of the obtained sample (Figure 1b), which could be ascribed to the flexible PETMP chains, as well as the various coordination structures between thiol/thiolate ligands and Zinc(II). TGA showed that the polymer was thermally stable below 300 °C (Figure 1c). The glass transition temperatures (Tg) of the PETMP80-20-Zn powder and glass were around 56 °C, according to tests performed using DSC (Figure 1d). The sample had a high elastic Young’s modulus of 1640 MPa, an elongation <5% before breaking ( Supporting Information Figure S7), and a flexural modulus of 1388 ± 87 MPa (n = 3, Figure 1e), which was about five times higher than the value of previously reported PDMS polymer crosslinked by strong covalent boroxine bonds.50 It could lift 500 g weight with just 1 mm thickness (Figure 1g). Moreover, only slight flexural strain (<2% before fracturing) was observed upon applying a flexural strength of 20 MPa, which confirmed the excellent rigidity of the material owing to the abundant coordination bonds between thiol/thiolate and Zn2+. Swollen experiments showed that this polymer was stable (no significant swelling or dissolution) in common solvents ( Supporting Information Figure S8 and Table S1). These features indicated that, at room temperature, our fabricated PETMP80-20-Zn CoAN behaved like a thermoset polymer in both mechanical strength and stability. Figure 1 | Synthesis and characterization of PETMP80-20-Zn. (a) Synthesis and processing of PETMP80-20-Zn. (b) PXRD of PETMP80-20-Zn. (c) TGA of PETMP80-20-Zn. (d) DSC thermogram of PETMP80-20-Zn. (e) The representative three-point flexural stress-strain curves for virgin samples and reprocessed samples. (f) Flexural properties of the virgin sample (cyan; 3 specimens tested) and reprocessed samples of PETMP80-20-Zn (blue; 3 specimens tested). (g) Optical images showing a plate sample with 20 mm in length, 10 mm in width, and 1 mm in thickness can withstand a load of 500 g. Download figure Download PowerPoint Reprocessability, reshapability, and healability Reprocessing, reshaping, and thermal healing ability are characteristic features of CANs and vital to sustainable industrial applications. For PETMP80-20-Zn, the out