Recyclable sulfur cured natural rubber with controlled disulfide metathesis | Communications Materials

Jun 29, 2025

Communications Materials volume 5, Article number: 212 (2024) Cite this article

3458 Accesses

4 Citations

14 Altmetric

Metrics details

Traditionally, sulfur-cured natural rubber compounds exhibit limited recyclability due to a significant drop in mechanical performance after reprocessing. Maintaining physical and chemical properties after recycling of a cross-linked polymer is an essential requirement for the global rubber industry to become more sustainable. Here, we demonstrate that tuning the curing process to favour a reversible cross-linked network based on disulfide and polysulfide bonds enables recyclability. We use a sulfur-based vulcanization system optimized with copper (II) methacrylate at concentrations of 2.47, 4.94, and 9.89 phr to control disulfide metathesis at low temperatures and enhance recyclability. Mechanical characterization identifies 2.47 phr as optimal for maintaining mechanical properties after initial moulding and full recovery after recycling. Additionally, we demonstrate that copper (II) methacrylate can be incorporated into existing rubber waste streams to promote recyclability.

In 2022 the annual consumption of natural rubber (NR) reached 14.975 million tonnes, while it is expected to reach 14.912 million tonnes for 20231. NR is widely utilized in a range of products across various different industries, including automotive, mining, construction, healthcare, and fashion, after undergoing cross-linking processes during manufacture. Among these applications, the largest consumption of NR is for tire production, accounting for more than 70% of the total global usage each year2. Consequently, the disposal of end-of-life tires (ELTs) presents a significant waste management challenge, with a substantial portion being deposited in landfill3. Accelerating towards a circular economy in the rubber industry necessitates establishing a connection between waste management, initial compound development, and product design and manufacture. This can be accomplished by simultaneously addressing each of these different aspects. The uncured rubber is essentially recyclable since it lacks cross-links, however it is unusable in most engineering applications. Cross-linking is predominantly introduced through a vulcanization process known as accelerated sulfur curing, whereby permanent covalent bonds form between the rubber polymer molecules, making recycling considerably more challenging (Fig. 1a). This sulfur-cured vulcanization process generates monosulfide, disulfide, and polysulfide cross-links between the polymer chains, with the ratio between them dependent on the sulfur/accelerator (S/A) ratio employed: a S/A » 1 is attributed to a conventional vulcanization system (CV) where 95% of the cross-links are polysulfides (and disulfides); with a S/A«1 an efficient vulcanization system (EV) is achieved where 80% of the cross-links are monosulfides; with S/A ≈ 1 a semi-efficient vulcanization system (SEV) is obtained where 50% of the cross-links are polysulfides4. Currently, significant effort is being made to recycle ELTs, focusing on either mechanical or chemical devulcanization processes that target carbon-sulfur and sulfur–sulfur bonds5,6. Unfortunately, these recycling methods typically yield degraded products, as the reprocessing mechanism lacks significant selectivity, and inevitably, this also affects main polymer chain carbon-carbon bonds, which produce materials with lower molecular weights and lower mechanical properties post-recycling. Despite the disadvantages, ELTs that undergo recycling often find application in fields requiring lower mechanical performance, such as for use as an additive in cement or pavements7. While there is extensive ongoing research into rubber waste management, more work is urgently required on how to develop manufacturing practices that can reduce rubber consumption and allow recycling. This could be done either by prolonging the lifetime of a cross-linked rubber under a specific application, for example, by developing self-healable rubbers8, or by designing a cross-linked rubber network that can be more readily recycled9,10. Disulfide bonds are known to be dynamic covalent bonds, which have been reported to undergo a free radical disulfide metathesis reaction11, that can be initiated by light or heat12 (Fig. 1b). It has also been established that they can follow a spontaneous disulfide metathesis13. These characteristics make them potentially useful for the development of sulfur-cured self-healable rubbers, as CV systems would introduce a higher proportion of disulfide and polysulfides as cross-links. However, in conventionally cured compounds that are characterized by a large number of polysulfides, once the sulfur-sulfur bond is cleaved and thiyl radicals are formed, they would statistically be keener to react with the NR chains, slowly shifting the cross-linked network to a SEV or EV system potentially with a higher crosslink concentration14,15. To prevent this and still exploit disulfide and polysulfides for the development of recyclable rubbers, metal-based inhibitors have been proposed. Examples of copper (II) chloride and copper (II) methacrylate (Fig. 1c) used on polybutadiene16 and polychloroprene17 rubbers, respectively, have previously been reported, reaching a recycling efficiency in terms of tensile strength of 55% for the former, and 94% in the latter.

a Sulfur curing of NR (in red) resulting in a three-dimensional cross-linked network held together with monosulphide, disulfide and polysulphide bridges (in green). b Disulfide metathesis following [2 + 1] free radical mechanism. c CuMA mediated disulfide metathesis. d Schematic representation for the three concentrations of CuMA (in green) used in this study, with respect to the number of disulfides (in black).

In the current work, we used an artificially simple unfilled compound to keep the chemistry as simple as possible. Copper (II) methacrylate (CuMA) was added to NR compounds cross-linked with a CV system at three different concentrations of 2.47 phr, 4.94 phr, and 9.89 phr, with the aim of inhibiting the disulfide metathesis at low temperatures, in order to preserve disulfide and polysulfide bonds and promote the rearrangement of the cross-linked network during reprocessing (Fig. 1d). The ratios 1/4, 1/2, and 1/1 reported reflect the molar ratios between CuMA and disulfides: with 2.47 phr of CuMa, a molar ratio between CuMA and disulfide of 1/4 was established; with 4.94 phr of CuMa, a molar ratio between CuMA and disulfide of 1/2 was established; with 9.89 phr of CuMa, a molar ratio between CuMA and disulfide of 1/1 was established. The use of CuCl2 was also considered, but preliminary work established that it is not compatible with NR18, as the prepared formulations slowly degraded after cross-linking. In sulfur vulcanization, the production of SO2 is inevitable, as is the presence of water in the environment. Although antioxidants can be used, CuCl2 is a mild oxidant that reacts with SO2 in the presence of water to produce CuCl, HCl, and H2SO4, which then react with and degrade the NR. The extent of degradation of rubber after initial compound mixing was evaluated using gel permeation chromatography (GPC), while optimal curing conditions were determined using moving die rheology (MDR). Recycling was performed by combining different ratios of reclaimed rubber and virgin (previously unvulcanized) rubber. The mechanical performance in pristine and recycled conditions was evaluated by quasi-static tensile testing. Additionally, the potential to recycle existing rubber waste was also evaluated by incorporating reclaimed CV Control samples with previously uncured CV CuMA formulations.

To demonstrate the chemical compatibility between CuMA and NR, GPC analysis was undertaken (Fig. 2a). When decomposition reactions take place, the molecular weight reduces (\({\bar{M}}_{{\mbox{w}}}\) of NR is 835,000 ± 42,000 g mol−1), and the molecular weight dispersity Ɖ increases (\({\bar{{{\mbox{-}} }{{D}}}}\) of NR is 3.35 ± 0.12). Additionally, during the mixing process, substantial mastication occurs, resulting in a decrease in the molecular weight of the NR chains. The addition of CuMA requires additional mixing time compared to CV Control formulation to ensure a complete dispersion in the rubber matrix. However, it was observed that the Ɖ values for the CV CuMA formulations were comparable to the values for the CV Control formulation (Table S1). This suggests that significant chain scission did not occur, as the molecular weight distribution was less dispersed in CV CuMA compounds compared to NR and CV Control compounds. To check the curing characteristics of CV systems, MDR analysis at 150 °C was carried out (Fig. 2a–d). The addition of CuMA results in lower reversion characteristics after reaching the maximum torque compared to the CV Control. Also, the final torque level decreases as the ratio of CuMA to sulfur-sulfur bonds is increased from 1/4 to 1/1. The higher torque values reported by CV CuMA 1/4 may be explained by a higher concentration of disulfide cross-links compared to CV Control, promoted through the coordination activity of CuMA. It could be speculated that the reduction in torque values reported by CV CuMA 1/1 and CV CuMA 1/2 could be attributed to a significant loss of polysulfide cross-links, in favor of generating more disulfide cross-links. Given that CuMA inhibits disulfide metathesis at low temperatures, while at higher temperatures allows sulfur–sulfur bond cleavage, it is possible that at 150 °C and under the cyclic shear strain in the MDR, there is a constant rearrangement of cross-links, assisted by a dynamic coordination established between Cu nuclei with the separated sulfur atoms. The evaluation of preliminary recycling potential was also carried out with MDR analysis by evaluating reversion control, scorch time, and torque recovery. All recycled CV CuMA compounds generally reported better reversion control compared to recycled CV Control compounds. Torque recovery improved significantly in all samples as soon as 10% of the uncured compound was incorporated, as, in fact, all R100V0 samples reported the lowest torque values. However, increasing the concentration of uncured compounds in recycled CV Control and recycled CV CuMA 1/4 compounds to 20% and 30% did not improve torque recovery, while it was observed for recycled CV CuMA 1/2 and recycled CV CuMA 1/1 compounds. Although in CV CuMA recycled compounds reversion is effectively managed, there is no improvement in scorch time recovery, indicating that the reclaimed compound does not exhibit the behavior of uncured rubber at 150 °C. The presence of a scorching time would be essential to demonstrate the full reversibility of the vulcanization process. Cross-link density evaluation through swelling experiments was carried out to further evaluate the recycling behaviors of CuMA-containing formulations (Fig. 3e). An increase in cross-link density is expected in recycled compounds as the reclaimed compound is mixed with uncured compound, and reprocessing is carried out at 150 °C for the same curing time equivalent to achieve 90% of the maximum torque, t90 as for the pristine cured compounds. As CuMA inhibits reversion during curing, the reclaimed portion of the recycled compound, which was previously cured to t90, during reprocessing reached t100, while the uncured portion of the recycled compound reached t90, resulting in an overall increase in the measured cross-link density. The cross-link density of recycled CV CuMA compounds effectively increases as the concentration of reclaimed compound is increased to 100% (Supplementary Table 2), from 1.35 × 10−4 mol g−1 (pristine) to 1.54 × 10−4 mol g−1 (R100V0) for CV CuMA 1/4, from 9.78 × 10−5 mol g−1 (pristine) to 1.46 × 10−4 mol g−1 (R100V0) for CV CuMA 1/2, and from 14.05 × 10−5 mol g−1 (pristine) to 8.31 × 10−5 mol g−1 (R100V0) for CV CuMA 1/1. By the same logic, the reversion in the CV Control compound results in reduced cross-link density in its recycled compounds as it drops for the R70V30 and then starts increasing with increasing concentration of the reclaimed compound. Additional microstructural changes can be evaluated by analyzing the Tg of the recycled compounds (Fig. 3f). Cross-linking inevitably leads to a small Tg shift towards higher temperatures as the polymer chains become somewhat constrained by the cross-linking reaction. The uncured NR used in this work had a Tg of −66.3 °C and the cured CV Control compound showed a shift towards a higher temperature of −59.2 °C, and a similar shift was reported by cured CV CuMA compounds, reaching −60.2 °C with 1/4, −61.7 °C with 1/2, and −61.8° with 1/1. During recycling, a further increase in Tg values is expected. However, the recycled CV Control and CV CuMA 1/4 compounds have two measurable Tg values for R70V30 and R80V20 samples (see Supplementary Fig. 2 for example). The higher Tg for these materials could be attributed to the reclaimed compound portion reaching higher cross-link density values, while the lower Tg could be attributed to the virgin compound portion that is now cross-linked. However, the lower Tg values are significantly lower than the one of the pristine cured compounds. This could be further explained by a probable sulfur migration from the uncured compound portion to the reclaimed compound portion, subsequently increasing the cross-link density in the reclaimed compound domains19,20. The ranking of uniaxial stress versus stretch ratio curves of pristine cured compounds is in accordance with the MDR torque and cross-link density results (Fig. 4). CV CuMA 1/4 has the highest stiffness and ultimate tensile strength value of 14.24 ± 1.11 MPa, followed by CV Control (11.24 ± 0.66 MPa), CV CuMA 1/2 (9.78 ± 3.33 MPa), and CV CuMA 1/1 (6.64 ± 0.39 MPa). For all formulations, recycling by blending different amounts of uncured compound seems to only affect the ultimate tensile strength and strain to break values, while no significant difference in stiffness is observed between R70V30, R80V20, and R90V10. Recycling of CV Control results in a reduction of stiffness, independently of how much-uncured CV Control compound is added (Fig. 4a). Recycling of CV CuMA 1/4 results in matching stress versus stretch ratio curves to the pristine cured compound, regardless of the quantity of uncured CV CuMA 1/4 compound added (Fig. 4b). The increase of CuMA to 1/2 (Figs. 4c) and 1/1 (Fig. 4d) results in an increase in the stiffness of the respective recycled compounds. To better understand the relation between stiffness, ultimate tensile strength, and strain to break values and uncured compound concentrations, the recycling efficiency in terms of ultimate tensile strength, ησ, and strain to break, ηε, was also evaluated (Fig. 5a–d). The obtained results are characterized by a wide standard deviation. This suggests that evaluation of recycling efficiency by comparison of ultimate tensile values is subject to a large experimental scatter, and this is therefore not a sensible measure of the recyclability of rubbery materials, in particular where values suggest that a recycling efficiency in terms of ultimate tensile strength over 100% were achieved, reflecting an increase in stiffness due to an increase in cross-link density, as seen for CV CuMA 1/2 and CV CuMA 1/1. However, some observations can be made on the dependency between ultimate tensile strength efficiency and concentration of uncured compound. In recycled CV Control and CV CuMA 1/1 compounds, ultimate tensile strength and strain efficiency increase with increasing concentration of uncured compound. The opposite behavior is observed for recycled CV CuMA 1/2 compounds, where ultimate tensile strength and strain efficiency values decrease with increasing concentration of uncured compound. Surprisingly, no trend is observed for recycled CuMA 1/4 compounds, as similar recycling efficiency values are obtained for all three R70V30, R80V20, and R90V10 samples, whilst they maintained the same stiffness. Finally, it is clear that recycling of 100% reclaimed compound is only achievable for CV CuMA 1/2 and CV CuMA 1/1 compounds, which can be associated with a disulfide-rich cross-linked network and a significant amount of CuMA, ensuring an efficient and controlled disulfide metathesis during recycling.

a GPC analysis results of \({\bar{M}}_{{\mbox{w}}}\) (in red), \({\bar{M}}_{{{\bf{n}}}}\) (in blue), and Ɖ (gray filled circles); error bars are reported in black; values reported in Supplementary Table 1. b MDR analysis at 150 °C for 60 min of CV Control (in black), CV CuMA 1/4 (in red), CV CuMa 1/2 (in green), and CV CuMA 1/1 (in blue); t90 values reported in Supplementary Table 1, repeatability of curves reported in Supplementary Fig. 1.

MDR analysis at 150 °C for 30 min of pristine cured compounds (in gray) and recycled compounds R100V0 (in red), R90V10 (in blue), R80V20 (in green), and R70V30 (in purple) of a CV Control, b CV CuMA-1/4, c CV CuMA-1/2, d CV CuMA-1/1. e Cross-link density of recycled compounds calculated from equilibrium swelling experiments of CV Control (black circles), CV CuMA 1/4 (blue circles), CV CuMA 1/2 (light blue circles), and CV CuMA 1/1 (aquamarine circles); error bars are reported in black; values reported in Supplementary Table 2. f Tg calculated from first derivative method applied to second heat run of DSC heat-cool-heat experiment of CV Control (black circles), CV CuMA 1/4 (blue circles), CV CuMA 1/2 (light blue circles), and CV CuMA 1/1 (aquamarine circles); error bars are reported in black; method for material with one Tg reported in Supplementary Fig. 2a and for material with two Tg in Supplementary Fig. 2b; values are reported in Supplementary Table 3.

Typical tensile test curves of pristine cured compounds (dotted black lines) and recycled compounds R100V0 (solid black line), R90V10 (solid blue line), R80V20 (solid red line), and R70V30 (solid green line), for a CV Control; b CV CuMA-1/4; c CV CuMA 1/2; d CV CuMA 1/1. Values are reported in Supplementary Table 4.

ησ results (in blue) and ηε results (in green) for a CV Control; b CV CuMA-1/4; c CV CuMA-1/2; d CV CuMA-1/1. Error bars are reported in black.

To understand whether CuMA can be incorporated in previously vulcanized rubber without CuMA, blends of 50% reclaimed CV Control compound and 50% uncured CV CuMA compound were prepared, and curing characteristics were evaluated (Fig. 6). The MDR analysis of the blended compounds reported a generally improved torque recovery compared to recycled CV Control compounds, particularly true when using uncured CV CuMA 1/4 (Fig. 6a). Additionally, the reversion in blended compounds is more effectively managed compared to pristine CV Control cured compound. Furthermore, there is partial recovery observed in scorch time compared to the one obtained from recycled CV Control compounds. The cross-link density of blended compounds reflects the torque recovery values, as the compounds with the lowest concentration of CuMA have higher cross-link density values (Fig. 6b). The blended compounds also reported two Tg values, similarly to some of the previously reported recycled compounds (Fig. 6c). In this case the lower Tg values are attributed to the uncured CV CuMA compounds incorporated in the blends and they are similar to the values reported for their pristine cured compounds counterparts. The higher Tg values, which are attributed to the reclaimed CV Control portions of the blends, all showed a significant shift towards higher values compared to the Tg value of pristine CV Control compound, of −59.2 °C, similar to the recycled CV Control compounds. The tensile properties of the blended compounds showed an improvement from the recycled CV Control compounds (Fig. 7a). All the blends showed the same stiffness, although this was lower when compared to the pristine CV Control cured compound. Additionally, from the recycling efficiency evaluation, it is clear that the ultimate tensile values gradually decreased with increasing concentration of CuMA from 1/4 to 1/1 (Fig. 7b). The reduction in strength with increasing concentration of CuMA is also observed in the pristine curves reported in Fig. 4b–d. CuMA promotes disulfide cross-linking rather than polysulfide cross-linking, and this usually translates into an increase in the modulus, but this is typically coupled with a reduction in overall strain and strength performance. Additionally, increasing the concentration of CuMA to the point of complete coordination of all disulfides would promote constant metathesis, but the reduction of polysulfides would inevitably lead to limitations in mechanical performance. The overall results suggest that recycling of previously vulcanized rubber without CuMA through blending with uncured rubber containing CuMA is possible, but optimal recovery is achieved with the lowest amount of CuMA reported in this study reaching ηε of 94% and ησ of 84% for CV CuMA 1/4 R80V20.

a MDR analysis at 150 °C for 30 min of blended compounds of reclaimed CV Control with uncured CuMA 1/4 (in red), with uncured CuMA 1/2 (in blue), and with uncured CuMA 1/1 (in green). All are compared to the MDR curve of pristine cured compound CV Control (in gray). b Cross-link density of blended compounds calculated from equilibrium swelling experiments (in black circles); error bars are reported in black; values are reported in Supplementary Table 5. c Tg calculated from first derivative method applied to second heat run of DSC heat-cool-heat experiment (in black circles); comparison is made with pristine cured compound CV Control (in dotted blue line); error bars are reported in black; values are reported in Supplementary Table 6.

a Typical tensile stress–stretch plots of blended compounds of reclaimed CV Control with uncured CuMA 1/4 (in red), with uncured CuMA 1/2 (in green), and with uncured CuMA 1/1 (in blue). All are compared to the stress-stretch curve of pristine cured compound CV Control (in gray); values are reported in Supplementary Table 7. b Recycling efficiencies in terms of stress (in blue) and strain (in green) for blended compounds; error bars are reported in black.

Recycling potential was evaluated through MDR analysis, examining reversion control, scorch time, and torque recovery. Notably, CV CuMA-1/4 exhibited controlled reversion within 30 min, a promising result compared to the CV Control. However, scorch time recovery is not achieved, indicating incomplete reversibility of vulcanization at 150 °C, as even though the cross-links are labile, a significant portion still shows some resistance reforming at the shear strain level in the MDR. Mechanical testing further supported the effectiveness of CuMA in recycling. Uniaxial tensile tests and stress versus stretch ratio curves illustrated the impact of CuMA concentration on the stiffness and stress-strain behavior of recycled compounds. CV CuMA-1/4 demonstrated matching stress versus stretch ratio curves compared with pristine compounds, emphasizing successful recycling. Recycling efficiency in terms of ultimate tensile strength and strain to break increased with higher uncured compound concentrations in CV Control. Blending reclaimed CV Control compound with uncured CV CuMA compounds demonstrated enhanced recovery, especially with lower CuMA concentrations. This study showcases the potential of CuMA in inhibiting disulfide metathesis and preserving essential cross-links for effective rubber recycling. It should be taken into account that because the data reported in this study is relative to pristine and recycled compounds that were processed and manufactured in laboratory conditions, the repeatability was affected by the quality of the reclaimed parts, eventual degradation processes, and by defects in the manufactured parts. Despite this, the results underscore the importance of balancing CuMA concentration to achieve optimal recycling efficiency, providing valuable insights into sustainable rubber manufacturing. However, the recycled and blended compounds reported in this study would be limited to applications following extrusion and compression molding only, given that the lack of scorch time excludes the possibility of injection molding. Additionally, the findings of this study are limited to CV rubbers only, therefore the applications would be limited to product requiring high strength at low temperature. Further research into optimizing the CuMA-based inhibition strategy by optimizing the reclaiming process, crumbling, and blending could help establish a way for a more robust and environmentally friendly rubber recycling process.

Natural rubber (NR, SMR CV60 grade) and n-cyclohexylbenzothiazol-2-sulfenamide (CBS) were purchased from the Tun Abdul Razak Research Centre. N-isopropyl-N’-phenyl-p-phenylenediamine (IPPD; 95% purity) and sulfur (~325 mesh, 99.5% purity) were purchased from Alfa Aesar. Copper (II) methacrylate (CuMA, technical grade) was purchased from Fisher Scientific. Stearic acid (StA; 95% purity, reagent grade) and zinc oxide (ZnO; ≥98% purity, Light Technical grade) were purchased from VWR International Ltd. Toluene (≥99.9% purity, HPLC grade) and tetrahydrofuran (THF; ≥99.9% purity, inhibitor-free, HPLC grade) were purchased from Honeywell International Inc.

In the present work, CV systems with and without CuMA were formulated and tested in different conditions before and after curing and recycling. The terminology adopted to refer to the formulations in specific conditions is as follows:

***Uncured compound: virgin rubber compound that has been mixed, but not cured.

Pristine-cured compound: newly cured compound.

Reclaimed compound: material that, after vulcanization, was collected for recycling.

Recycled compound: recycled cured compound.

Blended compound: recycled cured compound obtained by blending reclaimed CV Control compound with uncured CV CuMA compound.

Rubber formulations are reported in Table 1. NR was masticated in a laboratory two-roll mill for 5 min. The estimated time to incorporate each ingredient into the masticated rubber is reported in brackets. StA was incorporated first (10–15 min) to minimize roll sticking, reduce the mixture viscosity and hence improve the mixing process. Subsequently, ZnO was gradually added (20–30 min), followed by IPPD (20–30 min). Once a homogenous mixture was obtained, CuMA was gradually added (10 min for each gram of CuMA in 100 g of NR). Finally, CBS (5–10 min) and sulfur (10–15 min) were added, and the obtained uncured compound was molded in a manual hot press at 150 °C, applying a constant pressure of about 12 MPa, for a curing time equal to t90 determined using Monsanto Moving Die Rheometer (MDR) 2000 with the lower die moving at 1.66 Hz, which was different for each formulation (Table S1).

To recycle the pristine compounds, they were heated in the hot press for 5 min at 150 °C and immediately masticated in the two-roll mill until a coarse granulate was obtained. The material was then blended with the same uncured rubber in the following mass fraction ratios (“R” stands for “reclaimed” and “V” stands for “ virgin uncured”): 100% reclaimed rubber with no uncured rubber (R100V0); 90% reclaimed rubber with 10% uncured rubber (R90V10); 80% reclaimed rubber with 20% uncured rubber (R80V20); 70% reclaimed rubber with 30% uncured rubber (R70V30). The obtained blends were molded in the manual hot press at 150 °C for the same time equal to the t90 to obtain the related pristine compound (Table S1). For example: the time to mold a pristine CV CuMA 1/4 compound is 12 min, so the time to mold CV CuMA 1/4 R80V20 is also 12 min.

The number average molecular weight (\({\bar{M}}_{n}\)), the weight average molecular weight (\({\bar{M}}_{{\mbox{w}}}\)), and dispersity (Ɖ) of uncured compounds were determined using an Agilent Technologies 1260 Infinity GPC/SEC system equipped with a refractive index (RI) detector. Calibration was carried out using polystyrene standards (\({\bar{M}}_{{\mbox{w}}}\) range from 162 g mol−1 to 6,570,000 g mol−1) from Agilent Technologies, Inc. Samples were prepared at least 48 h prior to the analysis by dissolving approximately 10 mg of the sample in 5 mL of THF. The solutions were then filtered through a 0.2 μm PTFE syringe filter and transferred into GPC vials. Each run was performed by injecting 100 μL in the columns, kept at a constant temperature of 25 °C, with a flow rate of fresh THF of 1 mL/min for 40 min.

DSC was performed to determine the glass transition temperature (Tg) of pristine compounds and recycled compounds. The analysis was carried out using TA Instruments DSC25 equipment under a nitrogen flow. All samples were placed in Tzero Aluminum Hermetic pans, and the normalized heat flow was measured following heat–cool–heat experiments as follows: (1) first heating ramp from −90 °C to 250 °C at the rate of 5 °C/min; (2) second cool ramp from 250 °C to −90 °C at the rate of 5 °C/min; (3) third heat ramp from −90 °C to 250 °C at the rate of 5 °C/min. The Tg was calculated by evaluating the first derivative of the second heating curve, as reported in the supplementary information (Supplementary Fig. 2).

Equilibrium swelling experiments were carried out in toluene by placing a dry and pre-weighed rectangular shape sample for each cured compound in a vial containing excess toluene (approximately 0.2 g of rubber in 20 mL of Toluene). The vials were then left in a dark environment for 14 days. After this time, the sample was removed from the solution, tap-dried on a white lint-free tissue, and weighed again. The sample was then dried under reduced pressure to remove the solvent and then weighed again. Flory–Huggins theory was applied to calculate the cross-link density νphy of pristine compounds and recycled compounds using the following Eq. (1):

where ξ is the specific Flory–Huggins polymer-solvent interaction parameter, which is 0.385 for NR–toluene interaction, ν0 is the molar volume of the solvent, which is 106.28 cm3 for toluene, and VE is the volume fraction of elastomer, which can be calculated following Eq. (2):

where VS is the volume of the solvent absorbed, and VEN is the volume of the elastomer network, which can be determined using Eq. (3):

where mEN is the mass of the rubber and cross-linker used in the whole formulation, whose mtot is the total weight, m0 is the weight of the dried after-swelling sample, and ρE is the density of the sample, which was measured using a Micromeritics AccuPyc II 1345 gas-displacement pycnometer (number of purges 10, purge fill pressure 134.45 kPa, number of cycles 10, cycle fill pressure 134.45 kPa, equilibration rate 3.45 kPa min−1, chamber size 1 cm3).

Tensile testing until failure was carried out on pristine compounds and recycled compounds. Dumbbell shape samples were cut out using an ISO-37-4 die cutter 24 h after curing was completed. Testing was carried out using an Instron 68TM-10 machine equipped with a 2 kN load cell and pneumatic side action grips, using a rate of 1 mm s−1. The width and the thickness were measured prior to the start of the test, while the initial length, L0, was measured after gripping the sample, ensuring the width of the sample remained constant in the gauge length. Uniaxial stress, σ, strain, ε, and extension ratio, λ, were calculated using the following Eqs. (4)–(6) respectively:

where F is the measured force, A0 is the initial unstrained cross-sectional area, and ∆L the measured elongation through crosshead movement.

Recycling efficiency in terms of stress, ησ, was calculated by comparing the ultimate tensile stresses of pristine compounds and recycled compounds following Eq. (7):

The recycling efficiency in terms of strain, ηε, was calculated by comparing the ultimate tensile strains of pristine compounds and recycled compounds following Eq. (8):

GPC analysis was repeated three times. The experimental data shown is, therefore, the mean of the three values obtained, and the standard deviation is reported. DSC analysis was carried out three times. The experimental data shown is, therefore, the mean of the three values obtained, and the standard deviation is reported. Equilibrium swelling experiments were repeated three times. The experimental data shown is, therefore, the mean of the three values obtained, and the standard deviation is reported. Tensile tests in pristine, recycled, and blended compounds were repeated five times. The experimental data shown is, therefore, the mean of the five values obtained, and the standard deviation is reported.

All data (MDR, GPC, DSC, Tensile stress-stretch, Swelling) is available from the Queen Mary University of London open access repository, at the following URL: https://qmro.qmul.ac.uk/xmlui/handle/123456789/99346.

ANRPC revises up global natural rubber estimates. Eur. Rubber J. 10 https://www.european-rubber-journal.com/article/2094013/anrpc-revises-up-natural-rubber-supply-demand#:~:text=Citing%20preliminary%20data%20estimates%20from%20member%20countries%2C%20the,estimate%20announced%20by%20the%20industry%20body%20in%20May (2023).

Natural Rubber Supply Chain. EUROPEAN TYRE & RUBBER Manufacturers’ (2022). https://www.etrma.org/key-topics/materials/natural-rubber/. Accessed 30/01/2024.

Valentini, F. & Pegoretti, A. End-of-life options of tyres. A review. Adv. Ind. Eng. Polym. Res. 5, 203–213 (2022).

CAS Google Scholar

Gonzalez, L., Rodriguez, A., Valentin, J. L., Marcos-Fernández, A. & Posadas, P. Conventional and efficient crosslinking of natural rubber. KGK Rubberpoint 58, 638–646 (2005).

CAS Google Scholar

Alonso Pastor, L. E., Núñez Carrero, K. C., Araujo-Morera, J., Hernández Santana, M. & Pastor, J. M. Setting relationships between structure and devulcanization of ground tire rubber and their effect on self-healing elastomers. Polymers 14, 11 (2022).

Article CAS Google Scholar

Abbas-Abadi, M. S., Kusenberg, M., Shirazi, H. M., Goshayeshi, B. & Van Geem, K. M. Towards full recyclability of end-of-life tires: challenges and opportunities. J. Clean. Prod. 374, 134036 (2022).

Article CAS Google Scholar

Rigotti, D. & Dorigato, A. Novel uses of recycled rubber in civil applications. Adv. Ind. Eng. Polym. Res. 5, 214–233 (2022).

CAS Google Scholar

Nghiem, T. T., Nguyen, B. L., Huyen, L. T. & Kawahara, S. A novel approach to prepare self-healing vulcanized natural rubber using tetramethylthiuram disulfide. Polym. J. 55, 1097–1102 (2023).

Article CAS Google Scholar

Ye, J. et al. Self-healing and reprocessing of reclaimed rubber prepared by re-crosslinking waste natural rubber powders. Green Chem. 25, 6327–6335 (2023).

Article CAS Google Scholar

Trinh, B., Owen, P., vanderHeide, A., Gupta, A. & Mekonnen, T. H. Recyclable and self-healing natural rubber vitrimers from anhydride-epoxy exchangeable covalent bonds. ACS Appl. Polym. Mater. 5, 8890–8906 (2023).

Article CAS Google Scholar

Nevejans, S., Ballard, N., Miranda, J. I., Reck, B. & Asua, J. M. The underlying mechanisms for self-healing of poly(disulfide)s. Phys. Chem. Chem. Phys. 18, 27577–27583 (2016).

Article CAS PubMed Google Scholar

Kaur, A. et al. Novel crosslinking system for poly-chloroprene rubber to enable recyclability and introduce self-healing. Polymers 13, 3347 (2021).

Article CAS PubMed PubMed Central Google Scholar

Rekondo, A. et al. Catalyst-free room-temperature self-healing elastomers based on aromatic disulfide metathesis. Mater. Horiz. 1, 237–240 (2014).

Article CAS Google Scholar

Wang, M. et al. Effect of non-rubber components on the crosslinking structure and thermo-oxidative degradation of natural rubber. Polym. Degrad. Stab. 196, 109845 (2022).

Article CAS Google Scholar

Kong, Y., Chen, X., Li, Z., Li, G. & Huang, Y. Evolution of crosslinking structure in vulcanized natural rubber during thermal aging in the presence of a constant compressive stress. Polym. Degrad. Stab. 217, 110513 (2023).

Article CAS Google Scholar

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. 17, 4315–4325 (2015).

Article CAS Google Scholar

Xiang, H. P., Rong, M. Z. & Zhang, M. Q. Self-healing, reshaping, and recycling of vulcanized chloroprene rubber: a case study of multitask cyclic utilization of cross-linked polymer. ACS Sustain. Chem. Eng. 4, 2715–2724 (2016).

Article CAS Google Scholar

Chemical compatibility. Martin’s Rubber Company. https://www.martins-rubber.co.uk/technical-references/chemical-compatibility/. Accessed 30/01/2024.

Herrmann, V., Hanning, S. & Kreyenschmidt, M. Untersuchungen zur Diffusion des Schwefels in Rezyklat-Kautschukmischungen: teil 1: bestimmung des Diffusionskoeffizienten. GAK Gummi Fasern Kunstst. 71, 232–241 (2018).

CAS Google Scholar

Herrmann, V. et al. Untersuchungen zur Diffusion des Schwefels in Rezyklat-Kautschukmischungen: teil 2: schwefeldiffusion in Mischungen mit Gummimehl. GAK Gummi Fasern Kunstst. 71, 340–349 (2018).

CAS Google Scholar

Download references

This work was supported by the Engineering and Physical Sciences Research Council (EPSRC) grant EP/W018977/1.

School of Engineering and Materials Science, Queen Mary University of London, Mile End Road, London, E1 4NS, UK

Anureet Kaur, Meet M. Fefar, Thomas Griggs, Keizo Akutagawa & James J. C. Busfield

School of Mechanical and Aerospace Engineering, Queen’s University Belfast, Stranmillis Road, Belfast, BT9 5AH, UK

Biqiong Chen

Search author on:PubMed Google Scholar

Search author on:PubMed Google Scholar

Search author on:PubMed Google Scholar

Search author on:PubMed Google Scholar

Search author on:PubMed Google Scholar

Search author on:PubMed Google Scholar

All authors have read and approved the final manuscript. Anureet Kaur conceived the research idea and designed the experimental methodology, performed analysis and data interpretation, drafted the initial paper, and incorporated feedback from co-authors; Meet M. Fefar and Thomas Griggs contributed by conducting part of the laboratory experiments and collecting the relevant data; Keizo Akutagawa played a key role in discussing and refining the theoretical framework and implications of the findings; Biqiong Chen and James J. C. Busfield jointly secured research funding and reviewed the data and supported the interpretation; James J. C. Busfield supervised this work, and provided access to specialized equipment and facilities. All authors discussed the results and commented on the manuscript.

Correspondence to James J. C. Busfield.

The authors declare no competing interest.

Communications Materials thanks the anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available. Primary Handling Editors: Jet-Sing Lee.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

Kaur, A., Fefar, M.M., Griggs, T. et al. Recyclable sulfur cured natural rubber with controlled disulfide metathesis. Commun Mater 5, 212 (2024). https://doi.org/10.1038/s43246-024-00651-9

Download citation

Received: 04 March 2024

Accepted: 23 September 2024

Published: 06 October 2024

DOI: https://doi.org/10.1038/s43246-024-00651-9

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative