A Self‐Healing System for Polydicyclopentadiene Thermosets

Self‐healing offers promise for addressing structural failures, increasing lifespan, and improving durability in polymeric materials. Implementing self‐healing in thermoset polymers faces significant manufacturing challenges, especially due to the elevated temperature requirements of thermoset processing. To introduce self‐healing into structural thermosets, the self‐healing system must be thermally stable and compatible with the thermoset chemistry. This article demonstrates a self‐healing microcapsule‐based system stable to frontal polymerization (FP), a rapid and energy‐efficient manufacturing process with a self‐propagating exothermic reaction (≈200 °C). A thermally latent Grubbs‐type complex bearing two N‐heterocyclic carbene ligands addresses limitations in conventional G2‐based self‐healing approaches. Under FP's elevated temperatures, the catalyst remains dormant until activated by a Cu(I) co‐reagent, ensuring efficient polymerization of the dicyclopentadiene (DCPD) upon damage to the polyDCPD matrix. The two‐part microcapsule system consists of one capsule containing the thermally latent Grubbs‐type catalyst dissolved in the solvent, and another capsule containing a Cu(I) coagent blended with liquid DCPD monomer. Using the same chemistry for both matrix fabrication and healing results in strong interfaces as demonstrated by lap‐shear tests. In an optimized system, the self‐healing system restores the mechanical properties of the tough polyDCPD thermoset. Self‐healing efficiencies greater than 90% via tapered double cantilever beam tests are observed.


Introduction
[3][4] Traditional thermosets often require heating in autoclaves for full cure, resulting in high energy and capital expenditures. [4,5]Frontal polymerization (FP) [6][7][8] is an energy-efficient and rapid alternative, wherein a highly exothermic polymerization reaction self-propagates through a monomer resin at a steadystate front velocity ( f ) to provide a fully cured material.[11][12][13] At the polymerization front, monomers quickly transform into cross-linked networks.We have demonstrated that FROMP is compatible with a variety of manufacturing methods, ranging from additive manufacturing techniques (i.e., 3D printing) [14] to composite fabrication. [15][18] While p(DCPD) has a high fracture toughness, mechanical damage in service can still result from factors including thermal stresses and fatigue loading.Self-healing has potential to extend the service life of such thermosets by repairing damage, including fatigue cracks, as they occur.Self-healing is particularly attractive for mending cracks in difficult-to-access areas (e.g., outer space or underground) that otherwise require invasive and expensive repair or replacement operations. [19,20]Self-healing can generally be categorized into either intrinsic or extrinsic. [21]Intrinsic self-healing relies on molecularly driven processes that leverage reversible covalent bonds or covalent adaptive networks, such as transesterification, [22,23] cycloaddition, [24] and disulfide reshuffling reactions. [25,26]Despite its potential for multiple repair cycles and high self-healing efficiencies, challenges persist in sustaining the healing mechanism's chemistry over multiple cycles, and external inputs, such as thermal, UV, or chemical inputs, are often necessary to trigger the self-healing chemistry.Intrinsic self-healing predominantly suits nanoscale damage repair as it requires direct physical contact between fracture surfaces. [27]Extrinsic self-healing employs embedded microcapsules (MCs) or vascular networks to deliver liquid healing agents to the damage site, and self-healing occurs even when the fracture surfaces are not in direct contact.Because extrinsic self-healing does not require modification of the base material, it is compatible with many existing polymers. [28]It is important to note that for both intrinsic and extrinsic self-healing, when an external trigger is necessary to initiate the healing process, the self-healing should not be considered as "autonomic." Most prior literature on extrinsic self-healing has focused on the self-healing of low cure temperature epoxy and epoxy vinyl ester thermosets [29][30][31] and their composites [32,33] using microencapsulated (or microvascularized [34] ) DCPD and various Grubbs catalysts (e.g., [(SIMes)Ru(=CHPh)(PCy 3 )Cl 2 ]; G2).While ROMP-based self-healing has demonstrated a maximum 93% healing efficiency in the epoxy matrix at room temperature, [35] self-healing of a p(DCPD) matrix utilizing G2 or wax-coated catalyst particles has not been successful due to the thermal and chemical instability of these catalysts to the temperatures required to rapidly polymerize DCPD.Furthermore, limitations such as poor control over the size of the solid catalysts, inefficient mixing of the healing agents, and wax plasticization of the healed layer necessitate excessive amounts of catalysts (2.5 wt%) to achieve satisfactory healing efficiency and peak loads of 50 N, as measured in tapered double cantilever beam (TDCB) tests. [30]hermoset polymers such as epoxies, [36] polyimides, [37] silicon rubbers, [38] and p(DCPD) [39] often require higher curing temperatures to maximize their mechanical properties, e.g., fracture toughness and elastic modulus, attributes that are particularly vital for aerospace and other high-performance applications. [20,40]o impart self-healing to high-performance thermosetting polymers, the self-healing chemistry must survive the host matrix material's curing process, while the healing agents within the capsules and catalyst need to remain active at ambient temperatures.Previous attempts to improve the thermal stability of selfhealing components have only yielded moderate successes.Mangun et al. [41] introduced a poly(dimethyl siloxane)-based heal-ing chemistry for epoxy matrix, but curing at 177 °C yielded a self-healing efficiency of only 28%.Jin et al. [29] reported epoxyamine healing chemistries using a urea-formaldehyde (UF) and polyurethane (PU) double-layer microcapsule system designed to enhance the microcapsule thermal stability.Post-cure at 170 °C, however, reduced the healing efficiency to 61% due to the loss of the core materials.The effect of a polydopamine coating layer on a UF/PU shell was investigated [42][43][44] to increase the thermal stability and chemical resistance of solvent-filled microcapsules, and the coated microcapsules were stable at 180 °C for 2 h.However, solvent-based self-healing is limited to thermoplastic systems and is not appropriate for healing thermosets (which do not solvent weld).
Here we report a self-healing system for FROMP-cured p(DCPD) thermosets.We utilize a thermally stable self-healing chemistry in a dual-capsule system that remains active after exposure to the high temperatures reached in FROMP (≈ 200 °C).The stability is achieved through the use of a robust encapsulation system, a thermally resistant bis-N-heterocyclic (NHC) Rucarbene precatalyst, and a Cu(I) activation coreagent. [45]We note that because the same chemistry is utilized for matrix fabrication and self-healing strong bond formation to the fracture surfaces is observed.To the best of our knowledge, this is the first example of self-healing in p(DCPD).Leveraging this healing system, we successfully demonstrate self-healing of p(DCPD) thermosets prepared via FP, with healing efficiencies of 90% following optimization of microcapsule loading and healing agent stoichiometry.

Results and Discussion
Figure 1a illustrates a microcapsule-based self-healing system compatible with FROMP conditions (i.e., T ≈ 200 °C).The localized and rapidly propagating reaction wave in FROMP integrates self-healing microcapsules into the resulting p(DCPD) matrix.Incorporation of a rheology modifier (i.e., 5 wt% fumed silica) into the resin formulation prevents the segregation of microcapsules (Figure S1, Supporting Information).The introduction of fumed silica into DCPD specifically results in the solution exhibiting a yield stress ( y ).Rheological measurements reveal  y values of 3.8 and 63 Pa at 4 and 5 wt% fumed silica, respectively (Figure S2, Supporting Information).These  y are sufficient to inhibit gravitational sedimentation of the catalyst and monomer microcapsules.
The self-healing chemistry consists of a thermally latent Rubased metathesis catalyst in conjunction with DCPD monomer and Cu(I) activator as healing agents (Figure 1b).The bis-NHC complex, D899, is thermally stable and exhibits minimal background polymerization activity, [46] features unobtainable with G2.The D899 catalyst exhibits activity in ROMP only in the presence of a chemical activator such as Cu(I) compounds.This unique property ensures the catalyst's thermal stability and prevents premature activation during the high-temperature matrix fabrication process.
We are aware that incorporation of additives (e.g., self-healing microcapsules or rheological modifiers) into the FROMP system could affect the FP due to the added thermal mass, which acts as a thermal sink and reduces the embedded energy density released during cure.The frontal velocity ( f ) and frontal temperature (T max ) are measured for the resins containing different healing agent loadings (Figure 1c and Figure S3, Supporting Information).As a baseline,  f and T max of DCPD catalyzed by G2 (0.8 mm loading) without any additives are measured at 1.14 ± 0.04 mm s −1 and 214 ± 0.4 °C, respectively.Following the inclusion of a 5 wt% rheology modifier (fumed silica),  f remains consistent at 1.15 ± 0.05 mm s −1 , while T max decreases slightly to 199.5 ± 3.7 °C.Upon introduction of microcapsules at a loading of 5, 10, 15, and 20 wt%,  f exhibits reductions to 1.19 ± 0.06, 1.14 ± 0.06, 0.83 ± 0.10, and 0.66 ± 0.02 mm s −1 , respectively.Simultaneously, T max decreases to 195 ± 4, 187 ± 1, 177 ± 1, and 175 ± 2 °C, respectively.With the fumed silica and microcapsules, the FP front remains planar, and no thermal instabilities are observed (Videos S1 and S2, Supporting Information).The glass transition temperature (T g ) declines slightly with increasing microcapsule loadings, as depicted in Figure S4, Supporting Information.For pure p(DCPD) and p(DCPD) with silica-based rheology modifiers, the T g was measured at 130 °C.The introduction of 5, 10, 15, and 20% microcapsules led to T g decreasing to 126, 116, 113, and 107 °C, respectively.Similar drops in T g with increasing microcapsule concentration have been observed in prior works. [42,47]The reduction in T g suggests a potential reduction in the crosslink density within the thermoset network due to the presence of the additives.Figure S5, Supporting Information shows an increased swelling ratio in p(DCPD) containing microcapsules and rheology modifiers compared to unaltered p(DCPD).These findings emphasize the importance in identifying the optimal concentration of additives that minimally disturb the FP, as their incorporation may potentially impede or suppress the FP process.
To assess if microcapsule core material is lost from the capsules during or after FROMP, we prepare a thin strip of p(DCPD) containing 5, 10, 15, and 20 wt% of microcapsules between two glass slides (Figure 1d and Figure S6, Supporting Information).Optical microscopy reveals that the embedded microcapsules are void-free after FROMP, in contrast to a previous study [29] where the loss of the core contents was observed by the contrast in the microscope images.The microcapsules in the p(DCPD) matrix maintain their contents without leaking of the core phase at least for a month under typical storage conditions.After FP, the microcapsules release their core contents upon damage by razor blade as shown in Figure S7, Supporting Information indicating the DCPD remains unpolymerized.Thermogravimetric analysis revealed a reduction in mass above the boiling point of the microcapsule core matching closely the loading of microcapsules (Figure S8, Supporting Information).Finally, 6 months after sample fabrication the microcapsules in p(DCPD) still remain filled and void-free (Figure S9, Supporting Information).
We leverage D899 catalyst in the design of a self-healing system and optimize the ROMP activity of the D899 system to reduce temperatures required for self-healing.Previous studies [48,49] demonstrated that D899 remained inactive in the absence of an activator and resins with the 1.5 equivalent of P(O n Bu) 3 were observed to remain unpolymerized and viable for FROMP for 8 weeks without loss of activity.In the presence of a Cu(I) coreagent (e.g., CuCl), rapid Ru-to-Cu NHC transmetalation occurs to generate an active 14 e − metathesis complex, as depicted in Figure 2a.Previous studies [45,48] have demonstrated that various Cu(I) sources effectively activate D899 but at different rates, which influenced the overall degree of cure.The precatalyst remained inactive and unable to initiate ROMP in the absence of Cu(I), however, the activated catalyst was capable of initiating ROMP spontaneously (Figure 2b), which are desired properties for autonomous self-healing.
Resin formulations containing seven Cu(I) sources are evaluated (Figure 2c) and exhibit varying levels of ROMP reactivities, as measured by differential scanning calorimetry (DSC) postcure analysis.To evaluate the metathesis-based curing reaction of the two-part healing systems, the D899 latent catalyst is dissolved in DCPD (3.3 mM) and activated by adding the Cu(I) coreagent in DCPD (3.3 mm).Following quiescent storage at room temperature, the residual heat of polymerization (ΔH p ) of the resin blend is measured as a function of storage time.For lessactive coreagents such as Cu(PPh 3 ) 3 Br, the polymerization is extensively inhibited by the ancillary phosphine ligands, as previously reported. [48]Resins with this reagent remain in a liquid state and become a viscous liquid after a week.For Cu complexes bearing a single ancillary ligand (e.g., CuP(OEt) 3 I), catalyst activation occurs at ambient temperatures to partially polymerize the resin into a soft gel.Similar coordinatively unsaturated complexes bearing other ancillary ligands (e.g., Cu(SMe 2 )Br and [Cu(COD)Cl] 2 ) induce spontaneous polymerization upon mixing to form rigid p(DCPD) thermosets.In the case of simple salts of the form CuX (X = Cl, Br, I), exothermic polymerization reaction occurs spontaneously within minutes.Over time the residual heat of polymerization continues to decrease reaching 7.6 J g −1 after 8 days at ambient temperature.As a benchmark for practically complete cure, fully cured resins are prepared by curing the mixtures in an oven at 70 °C, resulting in residual ΔH p values between 3 and 20 J g −1 .The degree of cure is determined to be above 95% based on the value obtained from D899-derived formulation, which showed a ΔH p in the range of 250 to 350 J g −1 . [48]mong the Cu(I) coreagents, CuCl is selected for encapsulation based on its uniform mixing, reactivity, as well as compatibility with the encapsulation process.Considering the need for rapid mixing and initiation of polymerization during the healing event, highly active NHC acceptors (e.g., CuCl) without inhibiting ancillary ligands are the best candidates to mitigate reaction quenching.In contrast, other heavier isologous CuX species (X = Br, I) exhibit non-uniform mixing in DCPD resins.Complexes bearing ancillary species potentially degrade and release inhibiting ligands when exposed to acid during the encapsulation process.For this study, we find that CuCl provides the optimal rate of catalyst activation, and results in complete polymerization of the encapsulated DCPD self-healing agent.
For storability and long shelf-life of the self-healing components, a two-part microcapsule system was adopted to isolate monomer and catalyst in separate polymeric microcapsules.Even though the activity of the D899 exhibits ideal quasi-inert reactivity at ambient temperature, background polymerization occurs over long time spans (>2 months) even with an inhibitor. [48]To minimize the background reaction, solutions of D899 (11.7 mM) in phenylcyclohexane (PCH) are compartmentalized separately from DCPD monomer.For the activating coagent, mixtures of CuCl (3.3 mM) are encapsulated in DCPD (Figure 3a).We note that excessive PCH could result in the plasticization of the healed polymer layer which may be undesirable.To assess the impact of PCH on the mechanical properties of p(DPCD) polymerized in the presence of PCH, mechanical tests are conducted on p(DCPD) samples incorporating varying concentrations (0, 5, 10, 15, and 25 wt%) of PCH.PCH results in a reduction in Young's modulus and yield strength as the PCH concentration increases (Table S1, Supporting Information) with 25 wt% PCH resulting in the p(PCPD) exhibiting rubbery characteristics (Figure S10, Supporting Information).To assess if the impact of PCH on mechanical properties is permanent or transitory, the mechanical properties of p(DCPD) are evaluated after PCH removal via vacuum treatment at 40 °C for 4 days.Subsequent to PCH removal, there is an increase in both Young's modulus and yield strength to values exceeding 1 GPa and 40 MPa, respectively (similar to the properties of native p(DCPD).To minimize the quantity of PCH, small-diameter catalyst microcapsules containing high catalyst core concentrations are used.The high catalyst concentration reduces the total quantity of PCH and the small capsule size results in good distribution of catalyst throughout the sample.We suggest PCH present in the MCs diffuses into the surrounding matrix such that over time the healed layer becomes similar to the p(DCPD) after PCH removal.To understand how long this might take, we perform a series of experiments.To calculate the PCH diffusion rate into p(DCPD), we monitor the swelling ratio over time for p(DCPD) containing various silica and microcapsule loadings immersed in PCH (Figure S11, Supporting Information).Using this data, the diffusion coefficient for PCH in p(DCPD) is estimated to be 3.09 × 10 −8 cm 2 s −1 .A semiquantitative evaluation of PCH migration from the healed layer into the surrounding matrix indicates a 60 μm thick pure PCH layer will fall below 9% within 14 days (see Figure S12, Supporting Information).Since the initial PCH concentration in the selfhealing layer is not 100%, rather is at most 50 wt%, the actual PCH concentration in the healed p(DCPD) will be well below 5 wt% after 14 days.While it may be possible for PCH to evaporate, experimentally we find PCH evaporates very slowly from p(DCPD).A 5 mm thick slab of p(PCPD) containing 25 wt% PCH loses only 1-2% mass over 4 days at room temperature (≈22 °C), and even at 80 °C, the mass decrease is less than 7% over 2 days (Figure S13, Supporting Information).
We utilize in situ polymerization of the urea-formaldehyde (UF) prepolymer to encapsulate the healing agents via a modified literature procedure. [50,51]For the monomer microcapsule, an additional silica coating on the UF shell wall is employed to increase the monomer capsule's thermal stabilities (Figure 3b).We note that the catalyst and activating Cu complexes potentially undergo degradation or oxidation under encapsulation conditions.For example, D899 is known to undergo activation in the presence of HCl via NHC protonolysis. [46]Similarly, Cu(I) salts can oxidize during the emulsification process in an aqueous medium; the resultant Cu(II) species exhibit slower transmetalation rates with D899. [49]To mitigate these issues, neutral pH aqueous surfactant solutions are thoroughly deoxygenated prior to emulsification.In the case of CuCl encapsulation, added ascorbic acid (i.e., a reductant) minimizes adventitious oxidation during encapsulation. [52]fter synthesis, the capsules are filtered to remove unencapsulated materials and nanoparticulates, and then subjected to a spray-drying process producing a free-flowing powder.Aggregation of microcapsules during this process is avoided to prevent poor rupture capability at crack planes, where cracks may propagate between microcapsules instead of rupturing them.This process yields ≈ 100 g of material and is likely amenable to scale-up.The dried microcapsules retain their core phases and remain viable for self-healing for more than 3 months under ambient storage conditions.
Microcapsule morphology is observed by scanning electron microscopy (SEM), and size distributions are analyzed from optical micrographs (Figure S14, Supporting Information).Both varieties of microcapsules exhibit uniform Gaussian-like size distributions with average sizes of 8.6 ± 1.9 μm and 89.1 ± 9.3 μm for catalyst and monomer microcapsules, respectively (Figure 3d).Transmission electron microscopy (TEM) is used to determine the shell wall thicknesses of the catalyst (≈ 75 nm) and monomer (≈ 100 nm) microcapsules (Figure 3a,b, insets).Attenuated total reflectance-Fourier transform IR spectrum from pre-crushed microcapsules shows characteristic peaks associated with both the core solution and shell wall materials (Figure S15, Supporting Information).Elemental analysis, specifically, inductively coupled plasma mass spectrometry of the catalyst and monomer microcapsules indicates that the core contents contain Ru (0.07 wt%) and Cu (0.01 wt%), similar to that of the original core solution formulations.
Thermogravimetric analysis (TGA) is used to investigate the stability of the capsules under the temperatures observed during FROMP.At 200 °C, less than 1% of the original mass is lost (Figure 3c), suggesting the microcapsules and their payloads remain intact under FROMP-like conditions.The high boiling point of the core phases reduces the possibility of premature capsule rupture via pressure increases from vaporization.An additional silica coating on the UF shell wall increases the monomer capsule's thermal stabilities (Figure S16, Supporting Information).
The thermally stable self-healing microcapsules were included in FROMPed p(DCPD) to evaluate the system's self-healing performance (Figure 4a).][55] The localized short groove TDCB geometry employs a central opaque groove section which contains the test formulation incorporating self-healing microcapsules and rheology modifiers blended into the p(DCPD); the transparent surroundings consisted of p(DCPD) without additives, reducing the quantity of materials necessary for each test.
The fracture behavior of p(DCPD) without any additives exhibits rapid crack propagation after the tip opening of the precrack, leading to a sudden failure.Incorporation of rheology modifiers and microcapsules into the p(DCPD) matrix increases the matrix toughness and stabilizes the crack tip through local plastic deformation, thereby effectively arresting crack propagation and preventing catastrophic failure (Figure S17, Supporting Information).This behavior was reported previously for an epoxy matrix and primarily attributed to the crack pinning at microcapsules or particle sites. [56]igure 4b illustrates a representative load-displacement curve from virgin (black), healed (red), and fractured and unhealed (green) samples.The load-displacement curves of the healed and virgin samples plotted in Figure 4b illustrate the extent of recovery of the virgin mechanical properties.In the unhealed material, the slope of the curve was less than that of a healed sample.This indicates the fractured region of the sample is not capable of bearing a load as significant as the initial load. [53]o optimize the healing performance, various loadings of healing agents are evaluated (Figure 4c).The critical load, where deviation from the initial slope occurs and crack propagation starts, along with self-healing efficiency (′) calculated by the area under the load-displacement curves (see Supporting Information) are used to evaluate the self-healing performance. [30,35]The inclusion of silica-based rheology modifiers into DCPD resins toughens the matrix and increases the critical load from 240 to 350 N (Figure 4c).As healing microcapsules are incorporated into the matrix, the critical loads from the virgin samples decrease steadily, reaching 169 N for the samples containing 20 wt% microcapsules.For microcapsule loadings above 20 wt%, the mechanical properties decrease rapidly (but samples still self-heal).Following self-healing, the critical load increases from 128 N in composites with 5 wt% microcapsule loadings, reaching a maximum of 216 N for samples containing 15 wt% microcapsules.
SEM analysis of the fracture surface of the healed sample is used to characterize the healing behavior at the crack plane.As shown in Figure 4d, most microcapsules at the crack plane rupture and deliver healing agents into the crack plane.Healing agent polymerization then occurs, forming a new polymer layer on the fracture surface.A comparison between the crack plane before and after healing is presented in Figure S18, Supporting Information.In control experiments, solvents or DCPD monomer solution are injected into the crack plane via syringe to exclude the potential self-healing effect of solvent welding (Figure S19, Supporting Information).The solvent welding effect is not significant and solvent-weld samples exhibit a critical load of only ≈ 30 N.
Factors dictating the self-healing efficiency other than the quantity of healing agent released include catalyst-to-monomer ratio in the crack plane, solvent content in the healed layer, degree of polymerization, and interfacial bonding at the crack plane.The catalyst microcapsule to monomer microcapsules ratio is systematically varied to evaluate the effect of stoichiometry and solvent content on healing efficiency (Figure 5).If insufficient catalyst is released into the crack plane, polymerization is incomplete, and unpolymerized monomer evaporates or plasticizes the healed layer, which leads to poor healing efficiencies.At MC loadings lower than 15 wt%, a lack of healing agent at the fractured plane results in incomplete crack filling resulting in a low self-healing efficiency. [30]he self-healing efficiencies are calculated using the internal work (energy-to-failure) from virgin and healed fracture tests [30,35] (Equation (1) in the Supporting Information) due to the ductility of the load-displacement curves.The healing efficiencies are studied at fixed monomer microcapsule loading of 5 or 7.5 wt% with the remainder being catalyst microcapsules.Total microcapsule loadings above 15 wt% are not tested due to the resulting deterioration of the p(DCPD) mechanical properties.The increased loading of the catalyst microcapsules from 0.5 to 5 wt% at fixed 5 wt% monomer microcapsules results in gradual increases in healing efficiencies from 21.7 ± 9.6 to 68.1 ± 7.1%, respectively (Figure 5a).However, samples containing 5 wt% monomer capsules do not fully heal the crack; higher loadings of monomer capsules are required to achieve higher healing efficiencies.At a monomer microcapsule loading of 7.5 wt% and with catalyst microcapsule loadings from 1.9 to 7.5 wt%, the healing efficiency increases from 29.8 ± 3.2 to 90.8 ± 10.9%, respectively (Figure 5b).These results suggest that 7.5 wt% of monomer microcapsules with a 1:1 ratio of catalyst microcapsules provides the best selfhealing performance.A higher portion of the solvent (PCH) from the catalyst microcapsules did not appear to adversely affect the healing efficiency.As discussed, the PCH diffuses into the surrounding matrix, which ultimately leads to the formation of a rigid layer of p(DPCP) after healing.
To investigate the self-healing conditions at lower temperatures, an evaluation of self-healing performance at room temperature and near-ambient temperature (35 °C) is conducted as shown in Figure 5c.The healing efficiency is 39% after 1 week at room temperature and reaches 78% after 3 weeks.Self-healing at 35 °C showed a healing efficiency of 63% after 1 week and 84% after 3 weeks.The self-healing performance after 3 weeks is comparable to that obtained after 18 h at 80 °C, which suggests that at lower temperatures self-healing can reach full completion over longer times.
The incorporation of antioxidant (4,4′-methylenebis(2,6-ditert-butylphenol)) into the formulation improves the self-healing capabilities of the material. [57]In our system, both the matrix and healing agent are comprised of DCPD, and as such the bond between the matrix and the polymerized healing agent potentially is strong.][60] Incorporation of 2 wt% antioxidants into the DCPD resin prior to FP prevents polymer oxidation, and results in good inter-layer bonding as shown in Figure 5d.Lap-shear tests on samples with and without antioxidants are used to evaluate the layer adhesion strengths.The healing agents are injected between frontally polymerized p(DCPD) plates The embedded self-healing microcapsules show excellent longevity, as demonstrated by their effective performance for over a month exhibiting a self-healing efficiency of 94% (Figure S23, Supporting Information).This indicates the self-healing microcapsules remain functional in the p(DCPD) matrix over a prolonged period, showing their potential for long-term self-healing application.

Conclusion
We successfully demonstrate a self-healing system that survives high-temperature fabrication techniques such as during FROMP.A thermally latent bis-N-heterocyclic Ru-carbene catalyst is utilized to overcome existing issues in G2-based self-healing. [35]Utilizing this chemistry, self-healing performances of ≈ 91% are achieved.Notably, the self-healing chemistry described here requires lower loadings of catalyst (i.e., 1 wt% solution in the microcapsule core) and DCPD microcapsules (i.e., 7.5 wt%) compared to previously reported self-healing in epoxybased networks.The inclusion of antioxidants prevents delamination and debonding of the healed material and appears to be a key requirement for high self-healing efficiencies with DCPDbased systems.The microcapsules employed here remain viable for at least 3 months under typical storage conditions and survive the conditions experienced during FROMP of p(DCPD).The resultant thermosets exhibit self-healing capabilities for at least 1 month after fabrication.We suggest the system developed in this work provides an excellent platform to build on in future studies.

Experimental Section
Experimental details are shown in the Supporting Information.

Figure 1 .
Figure 1.Self-healing p(DCPD) fabricated by FP. a) General schematic of FP of self-healing p(DCPD).b) Thermally stable self-healing chemistry at the damaged site which repairs the cracks.c) Time-lapsed images of front propagation within initially liquid-state DCPD resins containing microcapsules.d) Optical micrograph of p(DCPD) containing catalyst and monomer microcapsules after FROMP.

Figure 2 .
Figure 2. Self-healing chemistry utilizing thermally latent catalyst and Cu(I) coreagent.a) In situ activation of D899 via NHC transmetalation to a Cu(I) coreagent unmasks a highly active metathesis catalyst.b) General polymerization schematic of DCPD monomer.c) Screening of various Cu(I) coreagents based on the residual heat of polymerization from DSC post-cure analysis.

Figure 3 .
Figure 3. Catalyst and monomer microcapsule characterization.Scanning electron microscope images of a) catalyst and b) monomer microcapsules.(insets) Transmission electron microscopy images of the microcapsule shell layer after microtome sectioning.c) Thermogravimetric analysis and d) size distribution of catalyst and monomer microcapsules.

Figure 4 .
Figure 4. Self-healing performance of microcapsule loaded p(DCPD).a) Tapered double cantilever beam (TDCB) test geometry for self-healing test.Microcapsules are embedded in the center region.b) Load-displacement curve of the virgin, healed, and unhealed samples.c) Critical loads where crack propagation starts from virgin (black) and healed (red) samples at different microcapsule loadings.The ratio between catalyst and monomer microcapsules is fixed at 1:1.The triangle point represents the fracture load from unadulterated p(DCPD) samples (no capsules).d) SEM image of the fractured surface of the TDCB specimen after healing.

Figure 5 .
Figure 5.Effect of microcapsule ratio on the self-healing efficiency.The self-healing efficiency is defined in the text.Effect of catalyst microcapsule loading on self-healing efficiency at fixed monomer microcapsule loading of a) 5 wt% and b) 7.5 wt%.c) Self-healing performance at room temperature and near-room temperature.The dashed line corresponds to the maximum healing efficiency after 18 h at 80 °C.d) Effect of antioxidant on lap shear performance of p(DCPD) prepared via FP.