Skip to main content
NCBI home page
JACS Au logoLink to JACS Au
. 2026 Jan 9;6(1):631–643. doi: 10.1021/jacsau.5c01642

Covalent Adaptable Networks from Commodity Polybutadiene and Rubber Waste

Daniel R Hart †, Nina B Georgieva †, Daniel M Krajovic ‡, Julia A Kalow †,*
PMCID: PMC12848713  PMID: 41614147

Abstract

Rubbers are polymer networks that are used in many everyday applications ranging from tires to apparel. Unfortunately, the cross-links that give these materials their desirable properties also make them difficult to recycle. Covalent adaptable networks (CANs) are a promising class of cross-linked polymers that rearrange their cross-links in response to a stimulus like heat, making them more recyclable than conventional thermosets. Herein we present a method of incorporating dithioalkylidenes, a catalyst-free associative dynamic bond, into polybutadiene rubbers using olefin metathesis. The modified polymers are cross-linked with a multiarmed thiol, and the resulting networks are chemically and mechanically recycled. Evolution of the network microstructure during recycling results in up to a 7-fold increase in toughness over three cycles of recycling. We incorporate common fillers like carbon fiber and silica into CANs to provide reinforced composites and recover these fillers through chemical recycling. Finally, we modify devulcanized rubber crumb derived from rubber waste, enabling the preparation of mechanically recyclable composites with 90% upcycled content. This work presents a new method of upcycling waste rubber to access materials with multiple lifecycles.

Keywords: vitrimer, covalent adaptable network, upcycling, recycling, rubber, elastomer

graphic file with name au5c01642_0009.jpg

graphic file with name au5c01642_0007.jpg

Introduction

Since the discovery of vulcanization, rubber has become ubiquitous for applications including tires, apparel, athletic equipment, and construction. Most commodity rubbers are vulcanized with sulfur or peroxides, which provides these materials with their characteristic robustness, elasticity, and strength. Unfortunately, this cross-linking process also makes it challenging to recycle rubber waste, with the United States Environmental Protection agency estimating that only 18% of rubber in the USA was recycled in 2018, while 54% was sent to landfills. , While programs in the EU have led to higher recycling rates (57% of tire materials recovered, 35% burned for energy recovery, and 8% in landfills or unaccounted for), current recycling methods primarily downcycle ground rubber crumb into secondary products such as pipes, shoes, or roofing, or use it as fillers for plastic composites. − This secondary recycling is preferable to landfilling or incineration; however, many of these products can only incorporate 15–50% recycled content without sacrificing material performance. Additionally, secondary recycling is not as sustainable as circularly recycling these materials back into a comparable product, or upcycling into a higher-value product. Therefore, it is crucial to identify alternatives to conventional vulcanization that offer opportunities for upcycling rubber and rubber waste.

One promising alternative to conventional thermosets is the class of polymers known as covalent adaptable networks (CANs). Like thermosets, CANs are three-dimensional polymer networks, but unlike permanently cross-linked thermosets, CANs are connected by dynamic covalent bonds, which allows them to rearrange under conditions that permit bond exchange. − This topological reconfiguration enables thermomechanical reprocessing and chemical recycling. CANs based on associative mechanisms are also known as vitrimers. To date, many types of associative dynamic covalent chemistries, including transesterification, , urethane − and vinylogous urethane exchange, − olefin metathesis, , silyl ether metathesis, − sulfonium transalkylation , and more have been used to create CANs.

Our lab has previously explored the use of dithioalkylidenes as cross-linkers in thiol-containing polymers to make CANs. Our work demonstrated that in the presence of free thiols, dithioalkylidenes undergo reversible conjugate addition–elimination of thiols without the need for an external catalyst, thus avoiding catalyst leaching and deactivation (Figure a). Additionally, we found that the stress relaxation of dithioalkylidene CANs can be tuned over several orders of magnitude simply by changing the structure of the carbon acid precursor. While these discoveries make dithioalkylidenes attractive candidates for CANs, the exchange mechanism previously confined these cross-linkers to use in specialty polymers that contain free thiols, such as thiol-modified PDMS and PEG-thiol. , To expand the potential applications of these dynamic cross-linkers, we explored methods to incorporate dithioalkylidenes into other high-volume commodity polymers, including rubber.

1.

1

Open in a new tab

(a) Mechanism of dithioalkylidene exchange in the presence of free thiols. (b) Metathesis allows dithioalkylidenes to be inserted into polybutadiene to prepare CANs, including surface functionalization of rubber crumb for reprocessable composites.

Guan first demonstrated that the double bonds in diene rubbers may be rearranged by olefin metathesis to enable malleability and self-healing, , but long-term recyclability requires sustaining the air-sensitive active catalyst or introducing new catalyst. , More commonly, researchers have incorporated dynamic bonds into rubber by epoxidizing polybutadiene (PB) or natural rubber (polyisoprene, PI), then ring-opening with bifunctional cross-linkers containing dynamic bonds. This method has been demonstrated with esters, ,− disulfides, , imines, and Diels–Alder adducts. However, in addition to raising the glass transition temperature (T g) of the rubber, , the residual epoxy groups undergo undesired side reactions upon aging and heating, compromising rubber properties. , Thiol–ene chemistry has also been reported as a facile method of introducing dynamic cross-linkers into rubbers, − and is currently being investigated in our lab. The associative dynamic bonds used in previous examples are boronic esters and imines, which could render these materials hydrolytically unstable. More importantly, while rubber crumb is often functionalized for use in secondary recycled products, , we found very few examples of converting waste rubber into CANs. ,, Furthermore, those examples require harsh conditions for reprocessing (160 °C for 1 h, 100 °C for 12 h, 180 °C for 3 h), which may also cause oxidative cross-linking. Therefore, we sought an alternative approach to incorporate associative dynamic bonds into diene polymers that would be directly applicable to waste rubber.

We hypothesized that strained cyclic alkene monomers could be incorporated into rubbers using a combination of olefin ring-opening and cross-metathesis (a variant of ring-opening insertion cross-metathesis polymerization, or ROIMP) (Figure b). In parallel with our studies, Foster et al. incorporated PB and acrylonitrile butadiene styrene copolymer into poly­(norbornene) through chain transfer of the diene polymers by cross-metathesis during ring-opening metathesis polymerization (ROMP). However, in their work, significant incorporation of norbornene comonomers produced thermoplastics with T g’s above room temperature, and cross-linking was not studied.

To harness ROIMP to prepare CANs from commodity polymers and waste rubber, we present a new norbornene monomer bearing a dithioalkylidene and show that it can successfully be incorporated into PB. Furthermore, we dynamically cross-link these modified copolymers, and chemically and mechanically recycle the resulting CANs with minimal changes to their tensile strength and increasing toughness. We hypothesize that this increase in toughness originates from changes to the polymer microstructure during reprocessing. Additionally, we prepare CAN composites with common polymer fillers such as silica and carbon fiber to show that chemical recycling of these composites enables filler recovery. Finally, this method is applied to devulcanized rubber particles to create new upcycled materials with a high level of recycled content.

Results and Discussion

Incorporation of Dynamic Bonds through ROIMP

Our first goal was to synthesize a new dithioalkylidene monomer capable of undergoing ring-opening metathesis (Scheme a). Such a monomer was prepared from cyclopent-4-ene-1,3-dione and cyclopentadiene, which react under ambient conditions to provide the previously reported endo-Diels–Alder adduct (DA) in greater than 95% yield without purification (see Supporting Information, Synthesis and Processing Procedures, for more details). The norbornene dithioalkylidene (ND) is then prepared by treating DA with potassium carbonate and carbon disulfide, followed by addition of methyl iodide to obtain ND in high yield and purity following purification (85% yield, Figure S2). To demonstrate the scalability of this synthesis, we carried it out on a 5 g scale and obtained ND in good yield (66%) and purity.

1. (a) Synthesis of ND; (b) Copolymerization of ND and COE to Produce p­(COE-co-ND); (c) ROIMP of ND and PB to Produce p­(B-co-ND) .

1

Open in a new tab

With ND in hand, we verified its ability to undergo ROMP. Due to the presence of two potentially chelating sulfurs in ND, there was some concern that this monomer might impede ROMP. While homopolymerization of ND proceeded slowly, as is typical for sterically hindered endo norbornenes, copolymerization of ND with cyclooctene (COE) using bench-stable second-generation Hoveyda–Grubbs (HG-II) catalyst was found to proceed readily (Scheme b). This copolymerization was performed with a 9:1 ratio of cyclooctene to ND and the reaction was allowed to proceed for 2 h at 23 °C to ensure full conversion. The polymer was analyzed using 1H NMR, which revealed that the final polymer composition matched the feed ratio of monomers, and the reaction time was similar to that of other endo norbornenes, indicating that the polymerization rate was not impeded by the presence of dithioalkylidenes (Figure S3). With no adverse effect observed for the HG-II catalyst in the presence of ND, we next turned our attention toward incorporation into commodity polymers.

PB rubbers are readily available and the second largest volume synthetic rubber after styrene–butadiene rubber (SBR); therefore, we chose to optimize ND incorporation into PB as a model polymer. , We dissolved high-cis PB (200–300 kg/mol, ∼2% 1,2-addition, MilliporeSigma) in CH2Cl2 (33% w/v) with ND (ca. 1 equiv per 25 PB repeat units) and HG-II (2 mol % relative to ND). Monitoring this reaction by 1H NMR showed that ND is completely consumed in the first hour, but if the polymer is collected and purified within the first 8 h, the collected products are found to be primarily unmodified PB with a small amount of p­(B-co-ND) with a high ND/butadiene ratio. This observation suggests that during the first few hours of the reaction, several processes including ND hompolymerization and polymer cross-metathesis occur concurrently. We discovered that significantly longer reaction times (∼24 h) and the addition of 1 mol % HG-II catalyst (relative to initial ND) after 18 h improved the random incorporation of ND, as the longer reaction times and extra catalyst enabled cross-metathesis to redistribute the ND through a greater fraction of the polymer chains. On a 10 g scale, the copolymer could be isolated by precipitation into acetone in 71% yield. (For experiments with other diene polymers, see Supporting Information, ND Incorporation into Other Polymers.)

The newly formed p­(B-co-ND) was initially characterized using gel permeation chromatography with UV–vis and refractive index (RI) detection (GPC) and NMR (Figures S4–S7). GPC chromatograms of the PB before and after metathesis showed that ROIMP had a significant impact on the final polymer’s molecular weight and absorbance. Prior to being treated with the metathesis catalyst, the commercial PB had a high molecular weight (M n = 120 kg/mol relative to polystyrene standards) and broad dispersity (D̵ = 4.0), which could only be observed using the RI detector (Figure S10). 1H NMR analysis confirmed that the commercial polymer is >95% cis. GPC analysis after ROIMP with ND shows that p­(B-co-ND) has a significantly lower molecular weight (M n = 7.2 kg/mol) and dispersity (D̵ = 1.8), which we found to be consistent with Watson and Wagener’s study on depolymerization of PB by metathesis catalysts (Figures S11 and S12). Even in the absence of solvent or chain limiters, the minority 1,2-addition repeat units promote depolymerization by ring-closing metathesis. Jones subsequently applied this depolymerization to deconstruct PB networks.

Incorporation of ND was confirmed by GPC and NMR. In contrast to commercial PB, GPC analysis indicated that the copolymer absorbed strongly at 360 nm, supporting incorporation of dithioalkylidenes (Figures S11 and S12). 1H NMR analysis confirmed that the sample contained both ND and butadiene repeat units and that the majority (∼80%) of the double bonds had isomerized from cis to trans in accordance with the previously observed thermodynamic ratio for metathesis catalysts (Figure S4).

In addition to GPC and 1H NMR, we further confirmed the coexistence of butadiene and ND repeat units in the same chains using diffusion-ordered NMR spectroscopy (DOSY NMR, Figure S5), which revealed the same diffusion coefficients for both components, as expected for a copolymer. Therefore, we conclude that true copolymers of ND and PB are formed, rather than separate homopolymers. It should be noted that the signals for PB diffuse over a wider range than those for ND, skewing toward lower diffusion coefficients. These broadened peak shapes could potentially indicate that lower-molecular-weight PB containing little to no ND is still present in the solution. Based on 1H and 13C NMR comparing the homopolymers and copolymer, we cannot conclusively determine whether the copolymer is random or blocky (Figures S6 and S7). Similar to Foster et al., we observe distinct signals in the copolymer that are not found in the homopolymers, which those authors attributed to heterodiads.

Dynamic Cross-linking and Microstructural Characterization

Having confirmed the successful synthesis of p­(B-co-ND), we next cross-linked the dithioalkylidene-functionalized polymer with pentaerythritol tetrakis­(3-mercaptopropionate) (PETMP) to make a CAN. We targeted 20% excess thiols relative to total binding locations (2 per ND, based on 1H NMR analysis) to enable network rearrangement via an associative mechanism. Networks were solvent cast by first mixing the copolymer with an antioxidant (to prevent oxidative cross-linking) and cross-linker in CH2Cl2. This mixture was stirred thoroughly, transferred to a Teflon sheet, cured for a few minutes at room temperature, and then cured under nitrogen and in vacuo at 80 °C. The films were then hot-pressed twice for 5 min at 120 °C to finish curing. This fabrication process resulted in a dark orange-brown network, net­(B-co-ND) (Figure a).

2.

2

Open in a new tab

(a) Image of net­(B-co-ND) after curing and pressing. (b) DMTA trace for net­(B-co-ND) showing multiple tan­(δ) peaks and a gradual drop in storage and loss modulus from −90 to 25 °C. (c) SAXS data for net­(B-co-ND) ranging from 50 °C to −60 °C with a maximum at q = ∼0.06 Å–1 (d) WAXS data for net­(B-co-ND) ranging from 50 °C to −60 °C, showing emergence of a sharp peak at q = 1.59 Å–1.

To determine the efficiency of the ROIMP as well as the curing procedure, we performed swelling experiments on the cured networks using CH2Cl2 and found that net­(B-co-ND) had a gel fraction of 71.4 ± 2.6% (Table S2). Analysis of the sol fraction by NMR showed that the soluble material was primarily PB homopolymer along with a small amount of antioxidant and other minor contaminants (Figure S8). These results confirmed the presence of nonfunctionalized PB chains that had previously been indicated by DOSY NMR, a problem that could potentially be mitigated in the future by using higher catalyst loading during ROIMP or longer reaction times to ensure more thorough incorporation.

The thermal properties of net­(B-co-ND) were characterized via thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC). Dynamic TGA showed that these materials experienced 5% mass loss at 302 °C, and when held at 120 °C these materials experienced less than 1% mass loss, demonstrating their stability at reasonable reprocessing temperatures (Figures S13 and S14). DSC revealed that this material is semicrystalline, as we would expect for materials with a significant fraction of trans-PB, with clear crystallization and melting peaks but no observable T g within the experimental temperature range of the instrument (Figure S15). Therefore, we turned to dynamic mechanical thermal analysis (DMTA) to evaluate the thermal transitions for this network. Instead of observing a rapid drop in the storage (E′) and loss (E″) moduli and a corresponding peak in the tan­(δ) trace at T g, the DMTA showed a gradual drop in storage and loss moduli ranging from −90 to 25 °C (Figure b and Supporting Information, Dynamic Mechanical Thermal Analysis), and the tan­(δ) displayed two broad overlapping peaks across the same range. Additionally, at temperatures above 50 °C, there was another gradual drop in the moduli, which was accompanied by a third peak in the tan­(δ) trace. We hypothesized that the gradual drop in moduli and the multiple tan­(δ) peaks arose from multiple thermal transitions caused by microphase separation and crystallization in net­(B-co-ND).

Variable-temperature small- and wide-angle X-ray scattering (SAXS and WAXS) were used to probe microscale phase separation and crystallinity (Figure c,d and Supporting Information, X-ray Scattering). At 50 °C, beyond the largest drop in modulus, only a broad peak corresponding to amorphous material is visible in the WAXS pattern, consistent with a fully amorphous molten matrix. The broad SAXS peak centered at ∼0.06 Å–1 indicates that the melt is microphase separated but lacks any order, , with a broad distribution of domain spacings that averages 9.7 nm, which are assigned as sulfur-rich domains (containing ND and the PETMP) in a PB-rich matrix. This observed disorganized microphase-separated morphology is reminiscent of our previous work on acrylic diblock copolymer CANs, as well as dioxaborolane-cross-linked polyethylene CANs. ,

Upon cooling to −8 °C, within the onset of the DSC exotherm, a sharp Bragg WAXS peak emerged at 1.59 Å–1, corresponding to the (200) plane of the monoclinic crystal structure of trans-PB and explicitly showing the growth of crystals in the matrix. Cooling further to −60 °C, below the peak crystallization temperature of the DSC exotherm, intensified the Bragg peak and signaled the completion of the crystallization process. Throughout these cooling stages, the SAXS peak diminished in intensity and shifted to slightly higher scattering vector. These features are consistent with crystallization reducing the electron density contrast between the PB- and sulfur-rich domains, reducing scattering intensity, and densifying the matrix to contract domain spacing. Heating to 50 °C reverses these processes, confirming that the DSC endotherm observed on heating is a melting transition and identifying matrix crystallization as a significant driver of the storage modulus changes occurring from 20 to −60 °C. We therefore assign the thermal transitions observed in the DMTA data to a combination of trans-PB crystallization and glass transitions of the PB-rich and ND-rich phases.

Stress Relaxation and Reprocessing

We next performed stress relaxation experiments to understand the reprocessability and stability of these CANs (Supporting Information, Stress Relaxation Data). Initial stress relaxation experiments were performed at temperatures ranging from 100 to 180 °C. These results showed that stress relaxation behavior became constant above 160 °C (Figure S28) due to significant levels of spontaneous cross-linking (Figure S29), a phenomenon that has previously been observed for PB at elevated temperatures even in the absence of sulfur or peroxides. , The presence of these permanent cross-links hinders stress relaxation and compromises reprocessability. Therefore, temperatures above 150 °C were avoided in future experiments.

Stress relaxation experiments, conducted in triplicate between 80 and 120 °C, revealed that the network can fully relax its stress within 100–1000 s (Figure a). These data were fitted to a stretched exponential model to find the characteristic relaxation time (τ*) and β value at each temperature (Table S8). The β values of ∼0.6 are consistent with an inhomogeneous, phase-separated material. The characteristic relaxation times were converted to average relaxation times, ⟨τ⟩, using a previously reported method. As expected for a cyclic dithioalkylidene, stress relaxation was rapid in these materials, with ⟨τ⟩ values ranging from 11 to 180 s across this temperature range. Using an Arrhenius plot (Figure b), the flow activation energy for this material was calculated to be 76.8 kJ/mol, which is in agreement with previously reported activation energies for dithioalkylidene exchange. , This flow activation energy greatly exceeds that of PB alone (33 kJ/mol) and confirms that relaxation is dominated by the exchange reaction. To rule out stress relaxation due to residual metathesis catalyst, , we prepared a control polymer by treating PB with HG-II without ND and cross-linking the resulting trans-majority PB using photoinitiated thiol–ene chemistry to provide a static thermoset network. Stress relaxation experiments at 120 °C for these control networks showed minimal stress relaxation even after 1 h (Figure S30), confirming that the rapid stress relaxation is due to thiol exchange with ND and not activity of residual metathesis catalyst.

3.

3

Open in a new tab

(a) Representative non-normalized stress relaxation data for net­(B-co-ND) ranging from 80 to 120 °C. All data are shown separately in Figures S31–S40. (b) Arrhenius fit of τ values obtained from stretched exponential fitting to the stress relaxation data. (c) Representative stress–strain curves of net­(B-co-ND) before and after recycling (strain rate of 1 mm/min at 23 °C). See Figures S43–S46 for replicates. (d) Observed changes to Young’s modulus and toughness before and after mechanical recycling. (e) Representative stress–strain curves of net­(B-co-ND) before and after recycling (strain rate of 1 mm/min at 23 °C). See Figures S48–S50 for replicates. (f) Observed changes to Young’s modulus and toughness before and after chemical recycling. All experiments were performed in triplicate.

Having established the potential for reprocessability via stress relaxation measurements, the mechanical properties of net­(B-co-ND) were evaluated before and after recycling (Supporting Information, Tensile Testing). Tensile tests were performed in triplicate on dogbone samples on a Testresources universal testing machine with a 220 N load cell. As initially prepared, net­(B-co-ND) displayed a Young’s modulus of 4.7 ± 0.2 MPa, tensile strength of 0.64 ± 0.1 MPa, and εb of 17 ± 3% (Figure c and Table S9). While the strength and extensibility of these networks is inferior to other examples of cross-linked polybutadiene in the literature (tensile strengths: ∼2–20 MPa; strain at break: 50–1000%), ,,, we hypothesized that much of the difference comes from the low molecular weight and high trans content of the p­(B-co-ND) following the metathesis reaction and not from the inclusion of ND. To test this hypothesis, we measured the tensile properties of the static PB control networks mentioned above. The control network also exhibited lower strength and extensibility than typical rubbers derived from cis-PB, with lower Young’s modulus and strength (0.53 ± 0.04 and 0.19 ± 0.04 MPa respectively) and higher εb (43 ± 10%) than net­(B-co-ND) (Table S9). These results confirm that the tensile properties of metathesis-modified PB materials primarily arise from the low molecular weight and high trans-content of the polymer backbone, and not from the dynamic bonds. In fact, the incorporation of ND strengthens the material, possibly by contributing physical cross-links in the form of stiff sulfur-rich domains. In the future, we anticipate that minimizing the decrease in molecular weight (by removing or reacting the 1,2-repeat units) and using cis-selective metathesis catalysts would likely result in improved mechanical properties for the network.

We then mechanically reprocessed net­(B-co-ND) by cutting and hot-pressing fragments at 120 °C and 10 tons ram force for 5 min. We measured the tensile properties following reprocessing over 3 cycles (Figure c,d). Over three cycles, we found that the tensile strength of net­(B-co-ND) was well maintained, ranging from 0.64 ± 0.1 to 0.76 ± 0.08 MPa. Surprisingly, while the material’s strength was constant, the εb and Young’s modulus changed significantly. The Young’s modulus decreased from 4.74 ± 0.2 to 1.65 ± 0.1 MPa, while εb increased from 17 ± 3% to 92 ± 13%. Together, these changes meant that the toughness of net­(B-co-ND) was significantly improved through reprocessing, increasing almost 7-fold from 60 ± 20 to 410 ± 97 kJ/m3. While identical properties are not observed after reprocessing, these materials can still be considered recyclable: the U.S. Environmental Protection Agency and European Environment Agency both define recycling as collection and processing/treatment of waste materials for reuse, with the latter explicitly specifying that material recycling can be accompanied by structural changes. ,

Like other CANs, dithioalkylidene networks can be dissolved by adding a large excess of monofunctional nucleophile (thiol), which pushes the equilibrium toward an uncross-linked state. With this in mind, we tested whether these materials underwent similar changes to toughness when chemically recycled with ethanethiol. Upon the addition of a large excess of ethanethiol, net­(B-co-ND) dissolved within 3–12 h at 23 °C. We noted an increase in dissolution time for samples that had been recycled multiple times, with new net­(B-co-ND) samples dissolving in about 3 h while the third cycle required 10–12 h for full dissolution. Due to the low boiling point of ethanethiol (∼35 °C), networks could be reformed without adding more cross-linker. Simply pouring the dissolved polymer solution directly onto a Teflon sheet and heating under nitrogen then vacuum pushed the equilibrium back toward the network state.

Surprisingly, the observed changes in tensile properties after each round of chemical recycling were almost identical to those observed during mechanical recycling (Figure e). The Young’s modulus once again dropped from 4.7 ± 0.2 to 1.8 ± 0.1 MPa, with somewhat smaller changes observed for the εb (16 ± 3% to 71 ± 10%) (Figure f). Tensile strength was still maintained between 0.64 ± 0.1 and 0.76 ± 0.09 MPa over 3 cycles. Due to the smaller εb changes observed during chemical recycling, a more modest ∼5× increase in toughness was observed across three cycles (60 ± 20 to 314 ± 87 kJ/m3).

To investigate the origin of mechanical changes upon both chemical and mechanical recycling, we repeated our characterization of samples after reprocessing. Since similar changes were observed for both reprocessing methods, we focused on characterizing mechanically recycled samples as this recycling process is simpler and faster. As shown in Figure a, DSC traces reveal significant changes in the melting and crystallization behavior of these materials after 3 cycles of recycling, with the crystallization peak increasing by 23 °C in the reprocessed materials. The melting temperature experiences a smaller shift, increasing by 8 °C. Furthermore, the crystallization and melting peak areas increase by 125%, indicating that reprocessing increases the crystallinity of these materials (Figure S16).

4.

4

Open in a new tab

(a) DSC data for initial and 3× mechanically recycled net­(B-co-ND) reveal changes in crystallization and melting temperatures. (b) DMTA data for net­(B-co-ND) as initially prepared and after 1, 2, and 3 rounds of mechanical recycling. (c) WAXS data from 50 °C to −60 °C for net­(B-co-ND) as initially prepared and after 3 rounds of mechanical recycling. (d) SAXS data for initial net­(B-co-ND) and mechanically recycled 3 times. (e) Cartoon schematic of hypothesized microstructural changes occurring in the net­(B-co-ND) during reprocessing which lead to an increase in toughness.

DMTA confirmed the shift in melting temperature that we had observed by DSC and, compared to as-prepared material, the recycled networks exhibit a decrease in storage modulus in the range of 30–60 °C, indicating a decrease in cross-link density. These observations are consistent with the decreased Young’s modulus measured through tensile testing and suggest that permanent cross-linking by adventitious thiol–ene and oxidative coupling are limited under the mild reprocessing conditions. Furthermore, the previously small thermal transition beginning between 40 and 50 °C became significantly more prominent after multiple rounds of recycling. In the initial net­(B-co-ND), there is a gradual drop in storage modulus and a tan­(δ) shoulder above 50 °C, but after several reprocessing cycles, a distinct transition is observed (Figure b). Since net­(B-co-ND) is not crystalline at this temperature (see WAXS data in Figures d and c), this transition is likely the glass transition temperature of the sulfur-rich phase. With the observed increase in crystallinity and more pronounced T g for the ND phase, we hypothesized that recycling net­(B-co-ND) coarsens phase separation.

To confirm that the change in mechanical properties was caused by coarsening of the phase separation, we performed variable-temperature X-ray scattering experiments. These experiments were performed on an as-prepared net­(B-co-ND) and a sample that had been recycled three times. As expected, our WAXS data showed that the material remained amorphous at temperatures above 50 °C for both samples, but upon cooling to room temperature and below, the Bragg peak emerges at 1.59 Å–1 with an increased sharpness and intensity for the recycled material (Figure c). By deconvoluting the Bragg peak and amorphous halo, we estimated an increase in crystallinity of ∼225% at room temperature and ∼113% at – 60 °C comparing reprocessed to as-prepared materials. However, it should be noted that the crystallinity of the materials was low at room temperature both before and after reprocessing (1.6–3.7%, Tables S4 and S5). After reprocessing, we also observed a significant shift in the broad SAXS peak (Figure d). As seen before, the initial net­(B-co-ND) exhibited a broad peak centered at 0.066 Å–1, or an average domain spacing of ∼9.4 nm, while the recycled net­(B-co-ND) peak appeared at 0.028 Å–1, corresponding to a far larger spacing of ∼22 nm. We hypothesize that this increase in domain spacing results in decreased effective cross-linking density as tie strands between sulfur-rich phases are converted to elastically ineffective loops, explaining the lower Young’s modulus observed by tensile testing and lower rubbery plateau modulus observed by DMTA. Increases in toughness and extensibility for net­(B-co-ND) likely arise from coarser phase separation following recycling (Figure e), which increases the crystallinity of the PB-rich phase and decreases interfacial area between the PB and sulfur-rich phases. This coarsening may also be responsible for an increase in stress relaxation time after recycling (Figures S41 and S42). We previously ascribed slow stress relaxation in microphase-separated vitrimers to the formation of isolated cross-link-rich domains, which hinders long-range chain diffusion. The relationship between crystallinity and ductility is not well understood in vitrimers; in a series of LLDPE-based vitrimers, the material with the lowest crystallinity and thinnest crystal lamellae was the most ductile, but highly crystalline samples of linear trans-PB homopolymers remain ductile well above T g. In our system, the increased crystallinity may enhance ductility if crystal deformation enables more net chain motion before network failure.

While the origin of these mechanical changes during chemical recycling are not understood, we speculate that the network does not fully dissociate into the p­(B-co-ND) prepolymers when dissolved, but rather forms soluble hyperbranched polymers that retain some of the network topology. Therefore, the ND-rich regions are retained, and are able to anneal together during cycles of dissolution, concentration, and heating.

Filler Incorporation and Recovery

Most rubber products are composites, with fillers like carbon black, carbon fiber, or fumed silica added to improve the material’s properties or decrease cost. To demonstrate that net­(B-co-ND) is compatible with some of these common fillers, we prepared composites using carbon fiber and fumed silica. Carbon fibers were incorporated into the networks both as a woven carbon fiber mat and as loose fibers. The woven mat was incorporated during the postcuring annealing process by layering two net­(B-co-ND) sheets on either side of the carbon fiber mat and hot-pressing to encapsulate the mat to make mC-net­(B-co-ND) (Figures a and S75). Because these CANs can be chemically recycled, we were able to dissolve the network with excess thiol and mild agitation to recover the carbon fiber by filtration. Composites with loose carbon fibers were prepared by simply mixing them into the p­(B-co-ND) solution prior to curing to create c-net­(B-co-ND). Tensile tests on c-net­(B-co-ND) revealed that the inclusion of loose carbon fibers significantly reinforced the network, with a Young’s modulus of 180 MPa and a tensile strength of 3.06 MPa (Figure b).

5.

5

Open in a new tab

(a) Carbon fiber mat before and after incorporation into net­(B-co-ND). See Figure S71 for additional photographs. (b) Representative stress–strain curves for c-net­(B-co-ND) and s-net­(B-co-ND) compared to net­(B-co-ND). Photographs of c-net­(B-co-ND) and s-net­(B-co-ND) are included near the appropriate curve. (c) IR data of ND, rubber crumb, and modified rubber crumb; the carbonyl stretch is highlighted. (d) Stress–strain curves for m-RC-X (X is weight % recycled content). For replicates, see Figures S51–S56. (e) Toughness as a function of recycled content; error bars are the standard deviation for 3 samples. For tabulated data, see Table S10. Inset: representative image of rubber films made from m-RC-90 (see Figure S74 for images of all samples). (f) Tensile strength of m-RC-X samples before and during 3 rounds of mechanical recycling. For stress–strain curves, see Figures S57–S66. For tabulated data, see Tables S11 and S12.

We also explored fumed silica as a reinforcing filler. Unlike carbon fiber, which is hydrophobic and could simply be added to the polymer during the curing process, unmodified fumed silica does not mix with PB. During conventional vulcanization, a silane coupling agent such as bis­[3-(triethoxysilyl)­propyl]­tetrasulfide (TESPT) is added to covalently bind the filler and rubber together via polysulfide bonds. To accommodate our exchange chemistry, which relies on free thiols as nucleophiles, we analogously functionalized silica with (3-mercaptopropyl)­methyldimethoxysilane (Figure S76). The resulting silica particles, now surface-functionalized with thiols, were then mixed with p­(B-co-ND) before curing, replacing the majority of the PETMP cross-linker, to create s-net­(B-co-ND). As expected based on the inclusion of filler and differences in cross-link density, tensile tests on s-net­(B-co-ND) also demonstrated significant reinforcement compared to unfilled net­(B-co-ND), with a Young’s modulus of 25 MPa, strength of 1.63 MPa, and strain at break of 10% (Figure b). When attempting to recollect the silica, we could not use simple filtration, as a stable suspension formed from s-net­(B-co-ND) when mixed with ethanethiol and CH2Cl2. Instead, the silica could be recollected through centrifugation in 85% yield. The yellow coloration of the recovered silica indicated that some of the polymer remained adsorbed to the surface; nevertheless, this residual polymer would not prevent the silica from being recycled into a new composite.

Upcycling Rubber Crumb

We next used the dynamic bond incorporation to upcycle rubber waste derived from truck tires. When scrap tires are collected for material recovery, they are shredded and the steel and fibers are removed. The resulting rubber crumb is a combination of carbon black and vulcanized polymers, including natural rubber (PI) and PB. We envisioned that the surface functionalization of ground rubber crumb would enable the particles to be dynamically cross-linked at their interfaces. Attempts to functionalize and cross-link vulcanized ground rubber crumb proved unsuccessful, possibly due to the inaccessibility of the PB alkenes to the metathesis catalyst. We obtained mechanically devulcanized rubber crumb from Tyromer (Ontario, Canada, estimated 75–85% devulcanized, Figure S72) and partially removed PI using Soxhlet extraction with hexanes, since its trisubstituted alkenes were previously shown to be unreactive. The PB remained strongly adsorbed to the carbon black and was not removed under any solvent conditions. The resulting rubber crumb (RC) was surface functionalized using HG-II and ND to provide modified rubber crumb (m-RC). Since RC and m-RC are insoluble in all solvents, we used IR to confirm the presence of dithioalkylidenes on the particle surface (Supporting Information, Fourier-Transform Infrared Spectroscopy). As seen in Figure c, RC exhibits no IR signals in the carbonyl region. m-RC shows a strong carbonyl peak at 1640 cm–1, corresponding to the carbonyl peak observed in pure ND, indicating successful modification of RC.

Gratifyingly, combining m-RC with PETMP followed by hot-pressing resulted in self-supporting cross-linked films (Figure S74). Tensile tests on m-RC films showed that the tensile properties were similar to s-net­(B-co-ND), with a Young’s modulus and strength of 38 ± 2 and 1.13 ± 0.08 MPa respectively, and a strain at break of ∼15 ± 4% (Figure d). Unlike net­(B-co-ND), m-RC films tended to break inhomogeneously, developing multiple cracks during a single tensile test. We hypothesized that these cracks arise due to inhomogeneity in the films caused by gaps between the m-RC particles. To create stronger, more homogeneous films, we fabricated a series of samples with 50–95% m-RC by mass, PETMP, and varying amounts of p­(B-co-ND) (m-RC-X , where X is the mass fraction of m-RC). The mechanical properties of these mixed films varied significantly, with intermediate amounts of recycled content (75–90%) showing the highest strength, extensibility, and homogeneity. Samples with 50 and 95% recycled content were weaker, more brittle, and less homogeneous, similar to the sample without added p­(B-co-ND) (Figure e). We hypothesize that sufficient added polymer is necessary to bridge gaps between m-RC particles and increase homogeneity, while materials with >25% p­(B-co-ND) suffer from the same mechanical properties as the unfilled CAN.

Since samples with 75–90% recycled content demonstrated the best materials properties, we mechanically recycled these composites at 80 °C for 35 min to test reprocessability. As shown in Figure f, after an initial drop, m-RC-90 maintains its strength through three rounds of reprocessing, while the strength of m-RC-85 and m-RC-75 decrease with each subsequent cycle (Figures S57–S66). Notably, m-RC-75 became so weak that it could hardly be loaded between the grips of the instrument before breaking. These experiments indicate that 90% upcycled content is optimal for both materials properties and recyclability, which is significantly higher than the 15–50% recycled content used in current applications, which are not themselves closed-loop. These materials did suffer from increasing inhomogeneity during multiple rounds of recycling on the hot press. Therefore, we anticipate that optimizing the reprocessing method (such as using extrusion or roll milling) could improve the homogeneity, and consequently their material properties, of the recycled samples over multiple cycles.

Conclusion

We have demonstrated that a new norbornene-based dithioalkylidene dynamic bond can be incorporated into commercial PB via olefin metathesis. The resulting copolymer is cross-linked through the addition of a tetrathiol, providing a CAN that maintains its strength and increases in extensibility through 3 cycles of mechanical and chemical recycling. While the properties degraded after 3 cycles, this performance still represents an improvement over conventional rubbers, which cannot be recycled at all. Through X-ray scattering, these changes were rationalized by the evolution of the network microstructure during reprocessing. Common fillers can be used to make dynamic reinforced composites from the modified polymer, and these fillers can be recovered by chemical recycling. Additionally, we have demonstrated that this method can be applied to functionalize the surface of devulcanized rubber crumb derived from tires. The resulting rubber crumb can be pressed into self-supporting films with a high recycled content, with optimal properties obtained using 90% modified crumb and 10% copolymer, which is a significantly higher percentage of recycled content than is currently used in most applications. This work opens a new pathway for making PB rubbers that can be recycled multiple times and shows promise for upcycling rubber waste into a new recyclable material with high-recycled content. In the future, we anticipate that the properties of these materials could be improved by better maintaining the molecular weight and cis content of the commercial PB, for example by using stereoretentive metathesis catalysts and removing the alkenes arising from 1,2-addition. While this initial study employed a halogenated solvent (CH2Cl2) for functionalization and cross-linking, in the future we anticipate that it could be replaced with green solvents such as 2-methyltetrahydrofuran, methyl ethyl ketone, or p-cymene, which are compatible with olefin metathesis. − Additionally, any large-scale implementation of this chemistry would necessitate the capture and reuse of thiols during curing and chemical recycling. We also intend to explore alternative processing methods to improve the homogeneity and reprocessability of the waste-derived materials.

Supplementary Material

au5c01642_si_001.pdf (7MB, pdf)

Acknowledgments

We thank Dr. Geoffrey Rojas at the University of Minnesota Characterization Facility for assisting with the setup of variable-temperature X-ray scattering experiments. We acknowledge Profs. Will Dichtel and John Torkelson for access to instrumentation, and Tyromer (Ontario, Canada) for the generous donation of devulcanized rubber.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/jacsau.5c01642.

  • Experimental and instrumental procedures, NMR spectra, size exclusion chromatography traces, and additional characterization data (PDF)

§.

Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge, MA 02139, USA

D.R.H. and J.A.K. conceptualized the project. D.M.K. performed the X-ray scattering experiments. N.B.G synthesized and characterized all mRC networks. D.R.H. performed the remaining experimental work and wrote the manuscript with J.A.K. The manuscript was written and edited with contributions from all authors. All authors have given approval to the final version of the manuscript.

This work was supported by funding from the National Science Foundation (NSF) Center for Sustainable Polymers (CHE-1901635) and partially funded by the Trienens Institute for Sustainability and Energy at Northwestern University. D.M.K. acknowledges funding from a University of Minnesota College of Science and Engineering Fellowship and an NSF Graduate Research Fellowship (DGE-2237827). This work made use of the CLaMMP Facility which has received support from the Northwestern University Materials Research Science and Engineering Center (NSF DMR-2308691), and the IMSERC Physical Characterization and NMR facility at Northwestern University (RRID: SCR_017874), which has received support from the Soft and Hybrid Nanotechnology Experimental (SHyNE) Resource (NSF ECCS-2025633) and Northwestern University. All X-ray scattering experiments were carried out at the Characterization Facility, University of Minnesota, which receives partial support from the NSF through the MRSEC (DMR-2011401) and NNCI (ECCS-2025124) programs.

The authors declare the following competing financial interest(s): J.A.K. and D.R.H. are co-inventors on a pending patent, US20250034341A1.

References

  1. Norton, M. G. RubberThe Material of Possibilities. In Ten Materials That Shaped Our World; Norton, M. G. , Ed.; Springer International Publishing: Cham, 2021; pp 125–144. [Google Scholar]
  2. US EPA . Rubber and Leather: Material-Specific Data. https://www.epa.gov/facts-and-figures-about-materials-waste-and-recycling/rubber-and-leather-material-specific-data (accessed April 01, 2025).
  3. Adhikari B., De D., Maiti S.. Reclamation and Recycling of Waste Rubber. Prog. Polym. Sci. 2000;25(7):909–948. doi: 10.1016/S0079-6700(00)00020-4. [DOI] [Google Scholar]
  4. Ricarte R. G., Shanbhag S.. A Tutorial Review of Linear Rheology for Polymer Chemists: Basics and Best Practices for Covalent Adaptable Networks. Polym. Chem. 2024;15(9):815–846. doi: 10.1039/D3PY01367G. [DOI] [Google Scholar]
  5. Europe92% of all End of Life Tyres collected and treated in 2017. https://www.etrma.org/wp-content/uploads/2019/11/20191119-Europe-92-of-all-End-of-Life-Tyres-collected-and-treated-in-2017.pdf (accessed July 14, 2025).
  6. Lapkovskis V., Mironovs V., Kasperovich A., Myadelets V., Goljandin D.. Crumb Rubber as a Secondary Raw Material from Waste Rubber: A Short Review of End-Of-Life Mechanical Processing Methods. Recycling. 2020;5(4):32. doi: 10.3390/recycling5040032. [DOI] [Google Scholar]
  7. Chittella H., Yoon L. W., Ramarad S., Lai Z.-W.. Rubber Waste Management: A Review on Methods, Mechanism, and Prospects. Polym. Degrad. Stab. 2021;194:109761. doi: 10.1016/j.polymdegradstab.2021.109761. [DOI] [Google Scholar]
  8. Jehanno C., Alty J. W., Roosen M., De Meester S., Dove A. P., Chen E. Y.-X., Leibfarth F. A., Sardon H.. Critical Advances and Future Opportunities in Upcycling Commodity Polymers. Nature. 2022;603(7903):803–814. doi: 10.1038/s41586-021-04350-0. [DOI] [PubMed] [Google Scholar]
  9. Denissen W., Winne J. M., Du Prez F. E.. Vitrimers: Permanent Organic Networks with Glass-like Fluidity. Chem. Sci. 2016;7(1):30–38. doi: 10.1039/C5SC02223A. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Podgórski M., Fairbanks B. D., Kirkpatrick B. E., McBride M., Martinez A., Dobson A., Bongiardina N. J., Bowman C. N.. Toward Stimuli-Responsive Dynamic Thermosets through Continuous Development and Improvements in Covalent Adaptable Networks (CANs) Adv. Mater. 2020;32(20):1906876. doi: 10.1002/adma.201906876. [DOI] [PubMed] [Google Scholar]
  11. Maes S., Badi N., Winne J. M., Du Prez F. E.. Taking Dynamic Covalent Chemistry out of the Lab and into Reprocessable Industrial Thermosets. Nat. Rev. Chem. 2025;9(3):144–158. doi: 10.1038/s41570-025-00686-7. [DOI] [PubMed] [Google Scholar]
  12. Montarnal D., Capelot M., Tournilhac F., Leibler L.. Silica-Like Malleable Materials from Permanent Organic Networks. Science. 2011;334(6058):965–968. doi: 10.1126/science.1212648. [DOI] [PubMed] [Google Scholar]
  13. Imbernon L., Norvez S., Leibler L.. Stress Relaxation and Self-Adhesion of Rubbers with Exchangeable Links. Macromolecules. 2016;49(6):2172–2178. doi: 10.1021/acs.macromol.5b02751. [DOI] [Google Scholar]
  14. Sun M., Sheppard D. T., Brutman J. P., Alsbaiee A., Dichtel W. R.. Green Catalysts for Reprocessing Thermoset Polyurethanes. Macromolecules. 2023;56(17):6978–6987. doi: 10.1021/acs.macromol.3c01116. [DOI] [Google Scholar]
  15. Kim S., Li K., Alsbaiee A., Brutman J. P., Dichtel W. R.. Circular Reprocessing of Thermoset Polyurethane Foams. Adv. Mater. 2023;35(41):2305387. doi: 10.1002/adma.202305387. [DOI] [PubMed] [Google Scholar]
  16. Kim S., Swartz J. L., Sala O., Sun M., Oanta A. K., Brutman J. P., Alsbaiee A., Dichtel W. R.. Thermal Activation of Zirconium­(IV) Acetylacetonate Catalysts to Enhance Polyurethane Synthesis and Reprocessing. Macromolecules. 2024;57(14):6759–6768. doi: 10.1021/acs.macromol.4c00625. [DOI] [Google Scholar]
  17. Sheppard D. T., Jin K., Hamachi L. S., Dean W., Fortman D. J., Ellison C. J., Dichtel W. R.. Reprocessing Postconsumer Polyurethane Foam Using Carbamate Exchange Catalysis and Twin-Screw Extrusion. ACS Cent. Sci. 2020;6(6):921–927. doi: 10.1021/acscentsci.0c00083. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Denissen W., Rivero G., Nicolaÿ R., Leibler L., Winne J. M., Du Prez F. E.. Vinylogous Urethane Vitrimers. Adv. Funct. Mater. 2015;25(16):2451–2457. doi: 10.1002/adfm.201404553. [DOI] [Google Scholar]
  19. Ma Y., Jiang X., Shi Z., Berrocal J. A., Weder C.. Closed-Loop Recycling of Vinylogous Urethane Vitrimers. Angew. Chem., Int. Ed. 2023;62(36):e202306188. doi: 10.1002/anie.202306188. [DOI] [PubMed] [Google Scholar]
  20. Denissen W., Droesbeke M., Nicolaÿ R., Leibler L., Winne J. M., Du Prez F. E.. Chemical Control of the Viscoelastic Properties of Vinylogous Urethane Vitrimers. Nat. Commun. 2017;8(1):14857. doi: 10.1038/ncomms14857. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Lu Y.-X., Tournilhac F., Leibler L., Guan Z.. Making Insoluble Polymer Networks Malleable via Olefin Metathesis. J. Am. Chem. Soc. 2012;134(20):8424–8427. doi: 10.1021/ja303356z. [DOI] [PubMed] [Google Scholar]
  22. Lu Y.-X., Guan Z.. Olefin Metathesis for Effective Polymer Healing via Dynamic Exchange of Strong Carbon–Carbon Double Bonds. J. Am. Chem. Soc. 2012;134(34):14226–14231. doi: 10.1021/ja306287s. [DOI] [PubMed] [Google Scholar]
  23. Tretbar C. A., Neal J. A., Guan Z.. Direct Silyl Ether Metathesis for Vitrimers with Exceptional Thermal Stability. J. Am. Chem. Soc. 2019;141(42):16595–16599. doi: 10.1021/jacs.9b08876. [DOI] [PubMed] [Google Scholar]
  24. Zych A., Pinalli R., Soliman M., Vachon J., Dalcanale E.. Polyethylene Vitrimers via Silyl Ether Exchange Reaction. Polymer. 2020;199:122567. doi: 10.1016/j.polymer.2020.122567. [DOI] [Google Scholar]
  25. Ash S., Sharma R., Shaker M., Patil S., Cheng S., Rabnawaz M.. One-Pot Synthesis of Robust Silyl Ether-Based HDPE Vitrimers with Enhanced Performance and Recyclability. Polymer. 2024;308:127374. doi: 10.1016/j.polymer.2024.127374. [DOI] [Google Scholar]
  26. Hendriks B., Waelkens J., Winne J. M., Du Prez F. E.. Poly­(Thioether) Vitrimers via Transalkylation of Trialkylsulfonium Salts. ACS Macro Lett. 2017;6(9):930–934. doi: 10.1021/acsmacrolett.7b00494. [DOI] [PubMed] [Google Scholar]
  27. Tang Z., Liu Y., Huang Q., Zhao J., Guo B., Zhang L.. A Real Recycling Loop of Sulfur-Cured Rubber through Transalkylation Exchange of C–S Bonds. Green Chem. 2018;20(24):5454–5458. doi: 10.1039/C8GC02932F. [DOI] [Google Scholar]
  28. Ishibashi J. S. A., Kalow J. A.. Vitrimeric Silicone Elastomers Enabled by Dynamic Meldrum’s Acid-Derived Cross-Links. ACS Macro Lett. 2018;7(4):482–486. doi: 10.1021/acsmacrolett.8b00166. [DOI] [PubMed] [Google Scholar]
  29. El-Zaatari B. M., Ishibashi J. S. A., Kalow J. A.. Cross-Linker Control of Vitrimer Flow. Polym. Chem. 2020;11(33):5339–5345. doi: 10.1039/D0PY00233J. [DOI] [Google Scholar]
  30. Zhang V., Accardo J. V., Kevlishvili I., Woods E. F., Chapman S. J., Eckdahl C. T., Stern C. L., Kulik H. J., Kalow J. A.. Tailoring Dynamic Hydrogels by Controlling Associative Exchange Rates. Chem. 2023;9(8):2298–2317. doi: 10.1016/j.chempr.2023.05.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Hemmingsen C. M., Chapman S. J., Deng C., Xiong Y., Hanley C. J., Zhang V., Olvera de la Cruz M., Kalow J. A.. Rheological Isotope Effects for Molecular Insight in Covalent Adaptable Networks. Macromolecules. 2025;58:7957. doi: 10.1021/acs.macromol.5c01258. [DOI] [Google Scholar]
  32. Lu Y.-X., Tournilhac F., Leibler L., Guan Z.. Making Insoluble Polymer Networks Malleable via Olefin Metathesis. J. Am. Chem. Soc. 2012;134(20):8424–8427. doi: 10.1021/ja303356z. [DOI] [PubMed] [Google Scholar]
  33. Cooper J. C., Paul J. E., Ramlawi N., Saengow C., Sharma A., Suslick B. A., Ewoldt R. H., Sottos N. R., Moore J. S.. Reprocessability in Engineering Thermosets Achieved Through Frontal Ring-Opening Metathesis Polymerization. Adv. Mater. 2024;36(28):2402627. doi: 10.1002/adma.202402627. [DOI] [PubMed] [Google Scholar]
  34. Xu Z., Witko M. L., Zheng H., Im J., Haber S., Ghosh A., Venetos M. C., Reimer J. A., Persson K. A., Helms B. A.. Resin-to-Resin Circularity in Chemical Recycling of Dicyclopentadiene-Based Cycloolefin Resins. J. Am. Chem. Soc. 2025;147(29):25613–25621. doi: 10.1021/jacs.5c06626. [DOI] [PubMed] [Google Scholar]
  35. Tang Z., Liu Y., Guo B., Zhang L.. Malleable, Mechanically Strong, and Adaptive Elastomers Enabled by Interfacial Exchangeable Bonds. Macromolecules. 2017;50(19):7584–7592. doi: 10.1021/acs.macromol.7b01261. [DOI] [Google Scholar]
  36. Qiu M., Wu S., Fang S., Tang Z., Guo B.. Sustainable, Recyclable and Robust Elastomers Enabled by Exchangeable Interfacial Cross-Linking. J. Mater. Chem. A. 2018;6(28):13607–13612. doi: 10.1039/C8TA04173C. [DOI] [Google Scholar]
  37. Wang J., Chen S., Lin T., Ke J., Chen T., Wu X., Lin C.. A Catalyst-Free and Recycle-Reinforcing Elastomer Vitrimer with Exchangeable Links. RSC Adv. 2020;10(64):39271–39276. doi: 10.1039/D0RA07728C. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Tong H., Chen Y., Weng Y., Zhang S.. Biodegradable-Renewable Vitrimer Fabrication by Epoxidized Natural Rubber and Oxidized Starch with Robust Ductility and Elastic Recovery. ACS Sustain. Chem. Eng. 2022;10(24):7942–7953. doi: 10.1021/acssuschemeng.2c01163. [DOI] [Google Scholar]
  39. Bui K., Wemyss A. M., Zhang R., Nguyen G. T. M., Vancaeyzeele C., Vidal F., Plesse C., Wan C.. Tailoring Electromechanical Properties of Natural Rubber Vitrimers by Cross-Linkers. Ind. Eng. Chem. Res. 2022;61(25):8871–8880. doi: 10.1021/acs.iecr.2c01229. [DOI] [Google Scholar]
  40. Kong L., Wu R., Zhang J., Huang S., Xie Z., Wu J.. High Performance, Self-Adhesive and Recyclable Dynamic Crosslinked Waste Rubber Particle Blends toward Upcycling of Waste Rubber. Green Chem. 2024;26(3):1523–1532. doi: 10.1039/D3GC03440B. [DOI] [Google Scholar]
  41. Zhu L., Xu L., Jie S., Li B.. Polybutadiene Vitrimers with Tunable Epoxy Ratios: Preparation and Properties. Polymers. 2021;13(23):4157. doi: 10.3390/polym13234157. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Zhu L., Xu L., Jie S., Li B.-G.. Preparation of Styrene–Butadiene Rubber Vitrimers with High Strength and Toughness through Imine and Hydrogen Bonds. Ind. Eng. Chem. Res. 2023;62(5):2299–2308. doi: 10.1021/acs.iecr.2c03133. [DOI] [Google Scholar]
  43. Berto P., Mehats J., Wirotius A.-L., Grelier S., Peruch F.. Reprocessable Covalent Elastomeric Networks from Functionalized 1,4-Cis-Polyisoprene and -Polybutadiene. Macromolecules. 2022;55(11):4557–4567. doi: 10.1021/acs.macromol.1c02156. [DOI] [Google Scholar]
  44. Heping Y., Sidong L., Zheng P.. Preparation and Study of Epoxidized Natural Rubber. J. Therm. Anal. Calorim. 1999;58(2):293–299. doi: 10.1023/A:1010199013587. [DOI] [Google Scholar]
  45. Chuayjuljit S., Yaowsang C., Na-Ranong N., Potiyaraj P.. Oil Resistance and Physical Properties of in Situ Epoxidized Natural Rubber from High Ammonia Concentrated Latex. J. Appl. Polym. Sci. 2006;100(5):3948–3955. doi: 10.1002/app.22998. [DOI] [Google Scholar]
  46. Gelling I. R., Morrison N. J.. Sulfur Vulcanization and Oxidative Aging of Epoxidized Natural Rubber. Rubber Chem. Technol. 1985;58(2):243–257. doi: 10.5254/1.3536063. [DOI] [Google Scholar]
  47. He C., Wang Y., Luo Y., Kong L., Peng Z.. Thermal Degradation Kinetics and Mechanism of Epoxidized Natural Rubber. J. Polym. Eng. 2013;33(4):331–335. doi: 10.1515/polyeng-2013-0040. [DOI] [Google Scholar]
  48. Chen Y., Tang Z., Zhang X., Liu Y., Wu S., Guo B.. Covalently Cross-Linked Elastomers with Self-Healing and Malleable Abilities Enabled by Boronic Ester Bonds. ACS Appl. Mater. Interfaces. 2018;10(28):24224–24231. doi: 10.1021/acsami.8b09863. [DOI] [PubMed] [Google Scholar]
  49. Yang Y., Huang L., Wu R., Niu Z., Fan W., Dai Q., Cui L., He J., Bai C.. Self-Strengthening, Self-Welding, Shape Memory, and Recyclable Polybutadiene-Based Material Driven by Dual-Dynamic Units. ACS Appl. Mater. Interfaces. 2022;14(2):3344–3355. doi: 10.1021/acsami.1c23007. [DOI] [PubMed] [Google Scholar]
  50. Ezzeddine D., Barzycki D. C., Ricarte R. G.. Impact of Cross-Link Density on Polybutadiene Permanent and Vitrimer Networks. ACS Macro Lett. 2025;14(7):1011–1018. doi: 10.1021/acsmacrolett.5c00267. [DOI] [PubMed] [Google Scholar]
  51. Breuillac A., Kassalias A., Nicolaÿ R.. Polybutadiene Vitrimers Based on Dioxaborolane Chemistry and Dual Networks with Static and Dynamic Cross-Links. Macromolecules. 2019;52(18):7102–7113. doi: 10.1021/acs.macromol.9b01288. [DOI] [Google Scholar]
  52. Trovatti E., Lacerda T. M., Carvalho A. J. F., Gandini A.. Recycling Tires? Reversible Crosslinking of Poly­(Butadiene) Adv. Mater. 2015;27(13):2242–2245. doi: 10.1002/adma.201405801. [DOI] [PubMed] [Google Scholar]
  53. Liu Y., Tang Z., Chen Y., Wu S., Guo B.. Programming Dynamic Imine Bond into Elastomer/Graphene Composite toward Mechanically Strong, Malleable, and Multi-Stimuli Responsive Vitrimer. Compos. Sci. Technol. 2018;168:214–223. doi: 10.1016/j.compscitech.2018.10.005. [DOI] [Google Scholar]
  54. Fazli A., Rodrigue D.. Recycling Waste Tires into Ground Tire Rubber (GTR)/Rubber Compounds: A Review. J. Compos. Sci. 2020;4(3):103. doi: 10.3390/jcs4030103. [DOI] [Google Scholar]
  55. Sampath M., Konikkara N., David S., S R., Rath J. P., P A. S. K.. Advancements in Sustainable Materials for Environmentally Responsible Tyre Production: A Comprehensive Review. RSC Sustain. 2025;3(10):4336–4363. doi: 10.1039/D5SU00177C. [DOI] [Google Scholar]
  56. Ye J., Tan S., Deng H., Huang W., Jin H., Zhang L., Xiang H., Zhang M.. Self-Healing and Reprocessing of Reclaimed Rubber Prepared by Re-Crosslinking Waste Natural Rubber Powders. Green Chem. 2023;25(16):6327–6335. doi: 10.1039/D3GC01652H. [DOI] [Google Scholar]
  57. Choi T.-L., Rutenberg I. M., Grubbs R. H.. Synthesis of A,B-Alternating Copolymers by Ring-Opening-Insertion-Metathesis Polymerization. Angew. Chem., Int. Ed. 2002;41(20):3839–3841. doi: 10.1002/1521-3773(20021018)41:20<3839::AID-ANIE3839>3.0.CO;2-H. [DOI] [PubMed] [Google Scholar]
  58. Foster J. C., Damron J. T., Zheng J., Guan C., Popovs I., Rahman Md. A., Galan N. J., Dishner I. T., Saito T.. Polyalkenamers as Drop-In Additives for Ring-Opening Metathesis Polymerization: A Promising Upcycling Paradigm. J. Am. Chem. Soc. 2024;146(48):33084–33092. doi: 10.1021/jacs.4c10588. [DOI] [PubMed] [Google Scholar]
  59. Clavier H., Giordano L., Tenaglia A.. Palladium-Mediated Phosphine-Dependent Chemoselective Bisallylic Alkylation Leading to Spirocarbocycles. Angew. Chem., Int. Ed. 2012;51(34):8648–8651. doi: 10.1002/anie.201204629. [DOI] [PubMed] [Google Scholar]
  60. Cater H. L., Balynska I., Allen M. J., Freeman B. D., Page Z. A.. User Guide to Ring-Opening Metathesis Polymerization of Endo-Norbornene Monomers with Chelated Initiators. Macromolecules. 2022;55(15):6671–6679. doi: 10.1021/acs.macromol.2c01196. [DOI] [Google Scholar]
  61. Kingsbury J. S., Harrity J. P. A., Bonitatebus P. J., Hoveyda A. H.. A Recyclable Ru-Based Metathesis Catalyst. J. Am. Chem. Soc. 1999;121(4):791–799. doi: 10.1021/ja983222u. [DOI] [Google Scholar]
  62. Burelo M., Yau A. Y., Gutiérrez S., Cruz-Morales J. A., Luna-Barcenas G., Treviño-Quintanilla C. D.. Metathesis of Butadiene Rubber for the Sustainable Production of Polyesters and Polyols. Polym. Degrad. Stab. 2024;227:110874. doi: 10.1016/j.polymdegradstab.2024.110874. [DOI] [Google Scholar]
  63. Friebe, L. ; Nuyken, O. ; Obrecht, W. . Neodymium-Based Ziegler/Natta Catalysts and Their Application in Diene Polymerization. In Neodymium Based Ziegler CatalystsFundamental Chemistry; Nuyken, O. , Ed.; Springer: Berlin, Heidelberg, 2006; pp 1–154. [Google Scholar]
  64. Watson M. D., Wagener K. B.. Solvent-Free Olefin Metathesis Depolymerization of 1,4-Polybutadiene. Macromolecules. 2000;33(5):1494–1496. doi: 10.1021/ma991560p. [DOI] [Google Scholar]
  65. Herman J. A., Seazzu M. E., Hughes L. G., Wheeler D. R., Washburn C. M., Jones B. H.. Depolymerization of Cross-Linked Polybutadiene Networks in Situ Using Latent Alkene Metathesis. ACS Appl. Polym. Mater. 2019;1(8):2177–2188. doi: 10.1021/acsapm.9b00434. [DOI] [Google Scholar]
  66. Peterson C., Hillmyer M. A.. Fast Photochromic Dye Response in Rigid Block Polymer Thermosets. ACS Appl. Polym. Mater. 2019;1(10):2778–2786. doi: 10.1021/acsapm.9b00741. [DOI] [Google Scholar]
  67. Bobrin V. A., Lee K., Zhang J., Corrigan N., Boyer C.. Nanostructure Control in 3D Printed Materials. Adv. Mater. 2022;34(4):2107643. doi: 10.1002/adma.202107643. [DOI] [PubMed] [Google Scholar]
  68. Ishibashi J. S. A., Pierce I. C., Chang A. B., Zografos A., El-Zaatari B. M., Fang Y., Weigand S. J., Bates F. S., Kalow J. A.. Mechanical and Structural Consequences of Associative Dynamic Cross-Linking in Acrylic Diblock Copolymers. Macromolecules. 2021;54(9):3972–3986. doi: 10.1021/acs.macromol.0c02744. [DOI] [Google Scholar]
  69. Ricarte R. G., Tournilhac F., Leibler L.. Phase Separation and Self-Assembly in Vitrimers: Hierarchical Morphology of Molten and Semicrystalline Polyethylene/Dioxaborolane Maleimide Systems. Macromolecules. 2019;52(2):432–443. doi: 10.1021/acs.macromol.8b02144. [DOI] [Google Scholar]
  70. Yang X., Cai J., Kong X., Dong W., Li G., Ling W., Zhou E.. Size-Affected Crystallization of Trans-1,4-Polybutadiene. Eur. Polym. J. 2001;37(4):763–769. doi: 10.1016/S0014-3057(00)00178-6. [DOI] [Google Scholar]
  71. Luttropp H.. The Thermal Vulcanization of Synthetic Rubber. Rubber Chem. Technol. 1958;31(1):132–146. doi: 10.5254/1.3542253. [DOI] [Google Scholar]
  72. Bellander M., Stenberg B., Persson S.. Crosslinking of Polybutadiene Rubber without Any Vulcanization Agent. Polym. Eng. Sci. 1998;38(8):1254–1260. doi: 10.1002/pen.10295. [DOI] [Google Scholar]
  73. Li L., Chen X., Jin K., Torkelson J. M.. Vitrimers Designed Both To Strongly Suppress Creep and To Recover Original Cross-Link Density after Reprocessing: Quantitative Theory and Experiments. Macromolecules. 2018;51(15):5537–5546. doi: 10.1021/acs.macromol.8b00922. [DOI] [Google Scholar]
  74. Ricarte R. G., Shanbhag S., Ezzeddine D., Barzycki D., Fay K.. Time–Temperature Superposition of Polybutadiene Vitrimers. Macromolecules. 2023;56(17):6806–6817. doi: 10.1021/acs.macromol.3c00883. [DOI] [Google Scholar]
  75. Cooper J. C., Paul J. E., Ramlawi N., Saengow C., Sharma A., Suslick B. A., Ewoldt R. H., Sottos N. R., Moore J. S.. Reprocessability in Engineering Thermosets Achieved Through Frontal Ring-Opening Metathesis Polymerization. Adv. Mater. 2024;36(28):2402627. doi: 10.1002/adma.202402627. [DOI] [PubMed] [Google Scholar]
  76. Short J. N., Thornton V., Kraus G.. Effect of Cis-Trans Ratio on the Physical Properties of 1,4 Polybutadienes. Rubber Chem. Technol. 1957;30(4):1118–1141. doi: 10.5254/1.3542749. [DOI] [Google Scholar]
  77. Xiang H. P., Qian H. J., Lu Z. Y., Rong M. Z., Zhang M. Q.. Crack Healing and Reclaiming of Vulcanized Rubber by Triggering the Rearrangement of Inherent Sulfur Crosslinked Networks. Green Chem. 2015;17(8):4315–4325. doi: 10.1039/C5GC00754B. [DOI] [Google Scholar]
  78. Khan R. K. M., Torker S., Hoveyda A. H.. Readily Accessible and Easily Modifiable Ru-Based Catalysts for Efficient and Z-Selective Ring-Opening Metathesis Polymerization and Ring-Opening/Cross-Metathesis. J. Am. Chem. Soc. 2013;135(28):10258–10261. doi: 10.1021/ja404208a. [DOI] [PubMed] [Google Scholar]
  79. US EPA . Recycling Basics and Benefits. https://www.epa.gov/recycle/recycling-basics-and-benefits (accessed Dec 03, 2025).
  80. European Topic Centre on Catalogue of Data Sources . Term: Recycling; European Environment Agency. https://www.eea.europa.eu/help/glossary/eea-glossary/recycling (accessed Dec 03, 2025). [Google Scholar]
  81. Flory, P. J. Principles of Polymer Chemistry, 19th print.; Cornell University Press: Ithaca, NY, 2006. [Google Scholar]
  82. Valdez S., Qiang Z.. Catalyst Influence on the Crystallization Behaviors and Mechanical Properties of Polyethylene Vitrimers. Macromolecules. 2025;58(21):11803–11815. doi: 10.1021/acs.macromol.5c02354. [DOI] [Google Scholar]
  83. van Aerle N. A. J. M., Lemstra P. J., Kanamoto T., Bastiaansen C. W. M.. Solid-State Deformation of Trans-1,4-Polybutadiene. Polymer. 1991;32(1):34–38. doi: 10.1016/0032-3861(91)90558-Z. [DOI] [Google Scholar]
  84. Xu S., Zhou J., Pan P.. Strain-Induced Multiscale Structural Evolutions of Crystallized Polymers: From Fundamental Studies to Recent Progresses. Prog. Polym. Sci. 2023;140:101676. doi: 10.1016/j.progpolymsci.2023.101676. [DOI] [Google Scholar]
  85. Ahn B., Lee J.-Y., Kim D., Kim I. J., Han S., Kim W.. Effects of Silane Agents and Curing Temperatures on Vulcanizate Structures. Rubber Chem. Technol. 2020;93(2):414–428. doi: 10.5254/rct.19.80445. [DOI] [Google Scholar]
  86. Valente M., Sibai A.. Rubber/Crete: Mechanical Properties of Scrap to Reuse Tire-Derived Rubber in Concrete; A Review. J. Appl. Biomater. Funct. Mater. 2019;17(1_Suppl):2280800019835486. doi: 10.1177/2280800019835486. [DOI] [PubMed] [Google Scholar]
  87. Product|Tyromer. https://tyromer.com/product/ (accessed July 25, 2025).
  88. Balashova I. M., Buduen R. G., Danner R. P.. Solubility of Organic Solvents in 1,4-Cis-Polybutadiene. Fluid Phase Equilib. 2012;334:10–14. doi: 10.1016/j.fluid.2012.07.030. [DOI] [Google Scholar]
  89. Bruneau, C. ; Fischmeister, C. . Olefin Metathesis in Green Organic Solvents and without Solvent. In Olefin Metathesis; John Wiley & Sons, Ltd, 2014; pp 523–535. [Google Scholar]
  90. Granato A. V., Santos A. G., dos Santos E. N.. P-Cymene as Solvent for Olefin Metathesis: Matching Efficiency and Sustainability. ChemSusChem. 2017;10(8):1832–1837. doi: 10.1002/cssc.201700116. [DOI] [PubMed] [Google Scholar]
  91. Shin H. G., Lee H. S., Hong E. J., Kim J. G.. Study of Green Solvents for Ruthenium Alkylidene Mediated Ring-Opening Metathesis Polymerization. Bull. Korean Chem. Soc. 2021;42(3):502–505. doi: 10.1002/bkcs.12213. [DOI] [Google Scholar]
  92. Qiao Z., Xu Q., Cheetham A. K., Jiang J.. High-Throughput Computational Screening of Metal–Organic Frameworks for Thiol Capture. J. Phys. Chem. C. 2017;121(40):22208–22215. doi: 10.1021/acs.jpcc.7b07758. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

au5c01642_si_001.pdf (7MB, pdf)

Articles from JACS Au are provided here courtesy of American Chemical Society

RESOURCES