Recycling of Carbon Fiber-Reinforced Epoxy Resin by Sustainable Solvolysis

Abstract

The increasing demand for epoxy-based carbon fiber-reinforced plastics (CFRPs) has led to significant waste generation and environmental concerns. To address this issue, a novel, recycling method has been developed using a mixture of methyl ethyl ketone (MEK) and hydrogen peroxide (H₂O₂) as a solvent system. This approach selectively decomposes the epoxy resin matrix under mild conditions, recovering carbon fibers with minimal damage. Optimal conditions for resin decomposition were identified at 70 °C for 12 h with a liquid–solid ratio of 15 in a MEK/H₂O₂/H₂O molar ratio of 1/1/4, achieving a decomposition ratio of up to 95%. The recovered carbon fibers were used to create new composites, which showed similar tensile properties to those reinforced with virgin carbon fibers. The decomposed resin was characterized using Fourier transform infrared spectroscopy (FTIR) and gel permeation chromatography (GPC), revealing a broad distribution of low-molecular-weight compounds. This study demonstrates a promising approach for the recycling of carbon fiber-reinforced plastics (CFRPs), which can help reduce waste generation and promote cost effectiveness and sustainability in the manufacturing of these materials.

1. Introduction

Epoxy-based carbon fiber-reinforced plastics (CFRPs) have recently migrated from the traditional aerospace and sports industries, where they have been extensively utilized for decades, to emerging applications such the automotive, energy, and construction sectors, driven by the decreasing cost of high-performance carbon fiber grades. Their advantageous properties, including corrosion resistance, durability, high heat resistance, light weight, high strength, and ease of fabrication, and the development of innovation, cost-reducing manufacturing methodologies have played a pivotal role in their successful introduction into these emerging sectors, leading to a significant escalation in their demand and consumption patterns. , However, the extensive utilization of CFRPs has led to significant material waste, including manufacturing offcuts and end-of-life parts, posing substantial environmental and sustainability challenge. Additionally, the high cost of virgin carbon fiber-reinforced composites (5 to 20 dollars per square meter) highlights the need for recycling. In order to recover carbon fibers from CFRP, it is necessary to remove the matrix resin that encases the carbon fibers. However, the main challenge in recycling CFRP is the strongly cross-linked network structure of the cured epoxy resin, which renders these materials difficult to dissolve due to their rigidity and three-dimensional stability.

The recovery CFRPs can be broadly categorized into three primary methods: mechanical, thermal, and chemical recycling. Mechanical recycling offers advantages in terms of low processing costs and ease of operation, but its limitations include the inability to recover long fibers and the restricted range of applications for shredded composites. Furthermore, the low value of the recycled products renders this approach wasteful, particularly when considering the high value of the original carbon fibers. Thermal recycling, on the other hand, has been widely adopted in various industrial sectors, with research focusing on two primary techniques: fluidized bed technology and pyrolysis. Fluidized bed technology enables the recovery of energy during the process, and the recovered fibers can be reused to manufacture new composites. Pyrolysis, however, allows for the retention of relatively high mechanical properties in recovered carbon fibers. Nevertheless, this approach is hindered by the presence of carbonized residue on the fiber surface and the emission of toxic gases during treatment, as well as the inability to recover the resin. Chemical recycling involves the degradation and subsequent dissolution of the resin in composites using solvents with or without catalysts so that the initial cross-linking is broken and the matrix can be dissolved to be removed, hence offering a promising approach for recycling of CFRPs. However, and despite some existing chemical recycling methods ensuring complete resin removal and recovery of long and continuous fibers with better strength, the use of hazardous chemicals and extreme processing conditions (supercritical fluids) limits the scalability of chemical recycling methods from both an environmental and economic perspective. Chemical recycling processes can be divided into three categories based on reaction conditions: supercritical fluid, subcritical fluid, and mild conditions. Although supercritical fluid recycling is a green alternative for chemical recycling of CFRP waste, it requires extremely high temperatures (400–500 °C) and pressures (10–30 MPa), resulting in high energy consumption. ,

Therefore, aiming mild processes and following a “green chemistry” perspective, hydrogen peroxide (H₂O₂) is highlighted as a promising oxidant for decomposing the epoxy matrix since water is the only byproduct. Unfortunately, its oxidization capacity is insufficient to degrade thermosetting resin effectively without an additional pretreatment. − Previous studies have shown that carbon fibers can be recovered using a variety of solvents, such as a nitric acid–H₂O₂ mixture under mild conditions. However, this process is not environmentally friendly due to release of small amounts of N₂O during the reaction and the difficulty to recover or dispose the spent acids. Other recent studies have shown that milder chemicals such acetic acid may replace nitric acid in these mixtures, achieving resin decomposition rates of up to 97% with 50% H₂O₂ concentration. , Furthermore, Petropoulos et al. reported a decomposition rate of up to 95% using an aqueous of 50 wt % H₂O₂ at 90 °C under ambient pressure conditions.

Building upon previous research, this study brings a novel strategy for efficient recycling of CFRPs under milder conditions and using environmentally friendly reagents; in particular, methyl ethyl ketone (MEK) is used instead of different acids such nitric or acetic acids. In combination with H₂O₂ as the oxidant for the decomposition of the epoxy resin, the recovery of carbon fiber has been evaluated by changing different conditions such temperature, concentration, MEK/H₂O₂ ratio, and reaction time, which can influence the quality of rCF (recycled carbon fiber). In addition, the obtained rCF was evaluated to be used as reinforcement in two different matrices: thermoset epoxy resin and thermoplastic polyamide 6 (PA6) matrixes, in order to produce new recycled composites, whose physicochemical properties are also evaluated.

2. Experimental Section

2.1. Materials

Expired CF-prepreg and CFRP offcuts were provided by Basque aeronautics company IDEC for chemical recycling. These materials were cut into 1 cm × 1 cm sections. The employed solvents were MEK (MEK, purity 99.93%) supplied by Scharlab and 50 and 30 wt % H₂O₂ supplied by Fisher Scientific.

For synthesizing the new composites with the rCFs, an aeronautical epoxy resin (RTM6) from Hexcel and a reactive thermoplastic matrix of PA6 were employed. Due to the low amount of rCF, both plates have been processed using a casting plus compression process in a closed mold (150 × 150 × 3 mm). The virgin carbon fibers employed in the comparison with the recycled carbon fibers were provided by SGL and SP System.

2.2. Experimental Procedures

The cut CF prepreg samples were immersed in a solvent mixture containing MEK and H₂O₂ of varying compositions at different concentration (15–30 liquid:solid ratio) into a reactor equipped with a reflux column and magnetic stirring. The reactor was heated to 50–70 °C for 6–24 h, depending on the reaction conditions. After treatment, the CF prepreg samples were separated from the solution, washed several times with ethanol, and then dried under vacuum at 60 °C for over 24 h. The resulting solution was evaporated under reduced pressure to recover the decomposed resin, which was then analyzed. The efficiency of matrix dissolution for different conditions was compared using the values of the resin decomposition ratio (D _R)

$$D_{\mathrm{R}}=\frac{(m_1-m_2)}{(m_1\times W_1)}\times 100$$ 1

where m ₁ is the mass of the initial composite, m ₂ is the mass of the dried sample after decomposition, and W ₁ is the mass fraction of resin in the initial composite, determined using ASTM 3171-09 standard.

The validation of the rCFs for their use as reinforcements in recycled composites was carried out by mixing them in an optimal ratio with the commercial resins, with curing during 2 h at 120 °C for epoxy resin and 10 min at 165 °C for PA6.

2.3. Characterization Techniques

Thermogravimetric analysis (TGA) was performed using the TGA/DSC 1 STARe System thermobalance (Mettler Toledo) to quantify the residual resin content on the carbon fibers. About 5–10 mg of sample was heated from 40 to 600 °C at a heating rate of 10 °C/min under a N₂ atm (50 mL/min).

Gel Permeation Chromatography (GPC) was used to determine the molecular weights of polymers, using a 1260 INFINITY II chromatograph, from AGILENT. Dimethylformamide (DMF) was used as the mobile phase. The average molecular weight of the decomposed resin was calculated using the calibration curve of polystyrene.

The equipment used for the identification of the functional groups of the decomposed resin was a Bruker FTIR spectrophotometer model ALPHA-P, equipped with the ATR Platinum accessory with a diamond crystal and a rebound that allows the analysis of all types of solid and liquid samples without sample preparation. The equipment has a spectral range of 7500–375 cm^–1 and a standard spectral resolution better than 2 cm^–1.

Surface and width of the virgin carbon fibers was compared with the recovered fibers using a high resolution Scanning Electron Microscope (HR-SEM, JEOL IT700HR) operated in high vacuum mode at an operating voltage of 5 kV, along with energy-dispersive X-ray spectroscopy (EDS) analysis.

The tensile tests of the recycled composites were carried out on a universal test machine INSTRON 34TM-50 (50 kN Load Cell) equipped with a contact extensometer INSTRON EX2620-602. UNE EN ISO 527-2:2012 was used as reference standard, modifying standard specimen dimensions and using rectangular specimens instead of classic bone-shaped ones. Modulus was calculated at 1 mm/min, and strength, when it was possible, was calculated at 2 mm/min.

3. Results and Discussion

3.1. Solvolysis of Epoxy Resin

Prior to testing, the initial CF prepreg and CFRP were characterized by TGA and ASTM 3171-09 in order to determine the initial fiber content and by SEM to examine the surface morphology of the samples. TGA results (Figure S1) showed a 40% of mass loss for CF-prepreg and a 30% for CFRP corresponding to the decomposition of the epoxy matrix, which is nicely aligned to the CF content determined by the ASTM standard of a 56% and 67% for CF-prepreg and CFRP, respectively. This observation is further supported by the SEM images, where CF-prepreg (Figure a) shows fewer visible CFs compared with CFRP (Figure b). It worth mentioning that due to the concordance of both techniques, D _R was determined by the application of the ASTM standard since it is quite more representative and accurate for composites due to the bigger sample size used for the analysis than that used for TGA.

1.

SEM images of virgin (a) CF prepreg; and (b) CFRP.

Once the initial composites were characterized, the effects of different reaction conditions, reaction time, MEK/H₂O₂/H₂O molar ratio, concentration, and temperature, were evaluated for an optimal carbon fiber recovery. As previously commented, due to the high cross-linking in epoxy resins, the most effective path to recover the fibers is the decomposition of the epoxy matrix, so a batch of different solvolysis conditions were tested, and the D _R results are shown in Table .

1. D _R of Composites under Different Reaction Conditions.

test	MEK/H2O2/H2O molar ratio	time (h)	temperature (°C)	solid/liquid ratio	D R	D R
#1	1/1/4	24	70	1:3	97	100
#2	4/3/16	24	70	1:3	92	95
#3	1/1/4	12	70	1:3	98	100
#4	1/0/0	24	70	1:3	32	68
#5	2/9/15	24	70	1:3	96	99
#6	2/3/5	24	70	1:3	96	100
#7	2/1/2	24	70	1:3	96	99
#8	0/2/3	24	70	1:3	89	93
#9	2/1/8	24	70	1:3	77	90
#10	2/1/2	12	70	1:3	91	95
#11	1/1/4	6	70	1:3	67	80
#12	2/1/2	6	70	1:3	68	80
#13	1/1/4	12	70	2:3	95	97
#14	1/1/4	12	50	2:3	27	63
#15	1/1/4	12	50	1:3	26	65

^aFrom TGA results.

^bFrom ASTM standard results.

The impact of different MEK/H₂O₂/H₂O molar ratios on the reaction extent was evaluated, revealing that the1/1/4 ratio exhibits the most positive outcome. Additionally, the similar results of D _R obtained for 2/9/15, 2/3/5, and 2/1/2 ratios (tests #5, #6, and #7, respectively) indicate that there is a range of different proportions in which the solvent system is useful. Moreover, single MEK and H₂O₂ yields D _R values of 32% (test #4) and 89% (test #8), respectively, so it is clear that there is a synergetic effect by mixing MEK and H₂O₂, hence highlighting it as a promising solvent system for resin decomposition. These notable effects can be clearly seen in Figure , in which carbon fibers are clearly separated from epoxy resin when using conditions of test #3.

2.

Pictures of recovery fibers collected after experiments (a) only with MEK (#4), (b) H₂O₂ (#8), and (c) MEK/H₂O₂/H₂O molar ratio 1/1/4 (#3).

Temperatures and times of reaction are crucial factors in the solvolysis of the epoxy resin, as shown in Table . Tests carried out at 50 °C led to D _R < 28% (tests #14 and #15), indicating a limiting reaction process. Increasing the reaction temperatures up to 70 °C provided energy enough to the system to enable the solvolysis process efficiently and to obtain D _R > 65% in most cases. On the other hand, the reaction time was found to significantly affect the decomposition process as well. Maintaining MEK/H₂O₂/H₂O molar ratios established in experiments #1 (1/1/4) and #7 (2/1/2), it was observed (Figure ) that reducing the reaction time to 12 h for the 1/1/4 ratio yielded a nearly complete decomposability ratio (D _R) of approximately 100% (#3), indicating that the reaction time can be optimized to 12 h without compromising process efficiency. However, a further reduction in time to 6 h (#11) resulted in a decreased D _R. In contrast, for the 2/1/2 ratio, shorter reaction times led to a more pronounced decline in D _R, with observed values of 91% (#10) and 68% (#12). This discrepancy may be attributed to the insufficient formation of potent oxidative species for the 2/1/2 ratio, highlighting the interplay between reaction time and oxidative species generation. Finally, the experimentation revealed that the liquid–solid ratio has a negligible impact on the solvolysis reaction, as evidenced by the consistent D _R of 95% and the lack of visual differences (Figure S2), hence suggesting that the ratio can be reduced without compromising the reaction efficiency, resulting in a substantial saving solvent.

3.

Effect of time on D _R 6 h (#11 and #12); 12 h (#3 and #10); and 24 h (#1 and #7) at a constant temperature and liquid/solid ratio (T: 70 °C, liquid/solid ratio: 30).

Therefore, it is clear that the ratio between the species within the solvent system is crucial for effective solvolysis of the epoxy resin. The results suggest that the most favorable outcome is achieved under the conditions employed in test #3, which yields optimal efficiency and effectiveness of the solvolysis. When considering the tests performed at 70 °C during 24 h (tests #1 to #9; excluding #3 with reduced reaction time), the experiments gave an optimal solvent ratio for MEK, H₂O₂, and H₂O, depicted as the red zone in Figure . Therefore, these results evidenced that there is a synergistic behavior in the decomposition system, characterized by the utilization of a mixed solution comprising H₂O₂ and MEK, which can be attributed to the in situ formation of dioxirane (Figure S3), a potent oxidizing species favored by the optimal 1/1 H₂O₂/MEK stoichiometric ratio. Moreover, the presence of excess water after the reaction supports this mechanism, because once the oxirane system has oxidized the epoxy resin until its decomposition, water is the byproduct of that reaction. Alternatively, the cross-linking components exhibit medium–high solubility in MEK, further influencing the system’s degradation behavior.

4.

D _R results of the solvolysis of the epoxy resin at different MEK/H₂O₂/H₂O molar ratios.

After the optimal process conditions were determined, the efficiency of the process was evaluated using a more complex residue type, specifically carbon fiber-reinforced plastic (CFRP), which exhibits a higher degree of cross-linking and reticulation due to its curing process. The results revealed a respectable D _R = 89%. For the CFRP composite, which is a multilayered solid, kinetic factors play a crucial role. Under these optimized conditions, the energy input is sufficient to enable the recovery of carbon fibers from the CF prepreg. However, for the CFRP residue, the radicals generated by H₂O₂ and MEK experience hindered access to the interior, resulting in a surface-initiated decomposition reaction that propagates inwardly. This is likely attributed to the fact that the CF prepreg residue is not really cured, although it is expired and therefore does not possess the same cross-linked structure as the CFRP residue, which has undergone a curing and postcuring cycle, leading to a chemically complex structure with a high degree of reticulation more difficult to oxidize.

3.2. Evaluation of Recovered Carbon Fibers and Oxidized Epoxy Resin

Surface morphology of recovered fibers was characterized by using a scanning electron microscope (SEM), as presented in Figure . After being treated, the micrographs reveal the presence of residual epoxy on the fiber surface, manifesting as discrete spots. Consistent with finding from other analytical techniques, CF prepreg exhibits a lower amount of resin residue compared to CFRP. Notably, the diameter of the recovered fibers, approximately 7 μm, is in good agreement with the value reported by Xu et al. for virgin fibers.

5.

SEM images of fibers recovered from (a) CF prepreg; and (b) CFRP.

The obtained resin after solvolysis (conditions of test #3) for both CF prepreg and CFRP were analyzed using Fourier transform infrared (FTIR) spectroscopy, as shown in Figure . The epoxy matrix of CFRP is a premixed epoxy-amine system, composed by the tetra-functional epoxy resin tetraglycidyl methylene dianiline and two hardeners: 4-4′methylenebis­(2,6-diethylaniline) and 4-4′-methylenebis­(2-isopropyl-6methylaniline) CFRPs not only show the absorption peaks for the amine cross-links represented by C–N bond stretch at 1100 and 847 cm^–1 corresponding to that of intact cross-links after the reaction but also absorbance at 1290 cm^–1 and 1350 cm^–1 corresponding to the nitro compounds resulting from the nitration of the main chain. Other peaks corresponding to the aromatic ring vibration (2900 and 1452 cm^–1), aryl alkyl ether, C–O–C (1240 cm^–1), epoxide functional groups (920 cm^–1), and the O–H stretch (3400 cm^–1) were also detected.

6.

FTIR spectrum of recovered resin from the composite CF prepreg and CFRP treated with experiment 13 conditions.

Regarding the CF prepreg matrix is a premixed epoxy-diamine system, composed by the tetra-functional epoxy resin tetraglycidyl methylene dianiline and two hardeners: 4,4′-diamino-difenil sulfona and triglycidyl-p-aminophenol. The spectra exhibit a prominent band around 1700 cm^–1, indicating the presence of carbonyl groups, which may be the product of hydroxyl oxidation. The disappearance of the 920 cm^–1 peak and the emergence of the 3220 cm^–1 peak suggest the ring-opening of oxirane and the subsequent formation of the secondary hydroxyl group. Additionally, there was also the 847 cm^–1 band that is assigned to C–N stretching.

Following this spectroscopic analysis, the underlying decomposition mechanism of the epoxy resin is proposed in Figure , which is in agreement with the hypothesis formulated by Das et al. In this scheme, R and R′ represent the –CH₂– groups, C–N bonds, and aromatic ring originating from the precursors (Figures S4 and S5). The synthesis process involves the establishment of amine cross-linking and the ring-opening of oxirane groups, resulting in the formation of hydroxyl groups. Oxidation of the resin subsequently leads to its decomposition, resulting mainly in the formation of nitro- and carbonyl functional groups. Complementarily, an alternative approach to elucidate the molecular composition of decomposed resin was performed by analyzing its molecular weight distribution using GPC (Figure S6). The analysis revealed a broad distribution of species with both extremely high and low molecular weights, with a predominance of low-molecular-weight compounds, hence evidencing a depolymerization, which is in concordance with an oxidative solvolysis caused by the MEK/H₂O₂/H₂O system in which the oxirane species are responsible for the oxidation.

7.

Proposed reaction mechanism for the decomposition of epoxy resin.

3.3. Verification of rCF Yield in Novel Composites

Carbon fiber recycled by solvolysis is indirectly validated mechanically by conducting mechanical tests on composite materials composed of recycled fibers and a selected matrix. This method has been employed in previous studies, as many mechanical testing equipment, such as tensile testing, require a minimum fiber length of 100 mm, which is challenging to achieve with fibers obtained from waste and recycled through a subsequent process.

For indirect validation, two matrices were selected, both of which have promising applications in the aeronautical and automotive sectors. The first matrix chosen was an epoxy resin, which is widely used in carbon fiber composites due to its outstanding properties. Specially, the RTM6 epoxy resin, used in the aeronautical sector through the liquid molding process, was studied in both the recycling and validation stages. The second matrix selected was an anionically obtained Polyamide 6 (PA6) thermoplastic, offering several advantages. As a thermoplastic material, PA6 can be easily melted and remolded, making it a sustainable alternative to thermoset. Additionally, PA6 meets the automotive industry’s increasing demand for thermoplastic composites that are easy to process and cost-effective.

Therefore, the recycled carbon fibers obtained in the solvolysis were used to prepare both recycled new composites. The resulting fiber percentages within the new composites were relatively lower than those typically used in the aeronautical industry (ca. 60 wt %) due to the difficulty of manufacturing recycled fiber fabrics that can withstand high-pressure impregnation processes and high fiber loadings at laboratory scale. However, both recycled and virgin fiber plates were manufactured by following the same process for comparison purposes.

The virgin fibers CF1 and CF2 used for comparison with rCF have some differences regarding their processing history. Although both fibers are virgin fibers, CF1 fibers are marketed in a Roving format, which allows the fibers to be aligned and grouped. Since rCF were separated into individual and twisted fibers during the recycling process, CF1 was carded to disrupt the fiber alignment and grouping, resulting in the CF2 fibers, morphologically more comparable to rCF. Figure shows the 5 manufactured composites following the procedure explained in Section , together with the weight percentage of CFs included in them.

8.

CFRP plates manufactured: (a) PA6/CF1 (19,01 wt %), (b) PA6/rCF (19,68 wt %), (c) Epoxy/CF1 (18,50 wt %), (d) Epoxy/CF2 (17,30 wt %), and (e) Epoxy/rCF (18,50 wt %). All plates have dimensions of 150 × 150 mm.

Comparison of the PA6 thermoplastic matrix plates shows a clearly lower modulus and strength for the rCF sample (Figure ). The decrease in modulus, which is mainly related to fiber quality and fiber orientation, indicates that the rCF has decreased in at least one of these two parameters. Initially it was thought that the rCF decreased in quality during the recycling process since both plates have a random fiber orientation. Tensile strength indicates the interaction between the fiber and the matrix. In this regard, it is important to point out that virgin CF comes with a specific surface treatment (sizing) that makes the matrix compatible with the fiber by creating a chemical anchorage between them. Recycled fiber loses this organic sizing during recycling, so the lack of this anchorage clearly justifies the decrease in the strength.

9.

Tensile test result for the 5 composite materials.

In the case of epoxy resin plates, 2 virgin fiber plates (CF1 and CF2) have been prepared to find out which of the above-mentioned effects has the greatest influence on the modulus. As already explained, Epoxy/CF1 and Epoxy/CF2 differed in fiber carding only for CF2. Figure shows a decrease in modulus from 12,900 to 9200 MPa, so it is concluded that the modulus decreases due to the difference between the roving and carding format. On the other hand, it is observed that the modulus remains the same between the Epoxy/CF2 and Epoxy/rCF samples, which confirms that there has been no fiber degradation due to chemical recycling. This conclusion could be transferred to the result of the PA6 matrix samples. Finally, if the strength results are analyzed, it can be seen that the value is maintained between the two specimens with virgin fiber, which is justified by the correct fiber–matrix interaction thanks to the specific sizing. After chemical recycling, in which the sizing is removed, the Epoxy/rCF specimen decreases its strength, as expected.

4. Conclusions

This study presents a novel and environmentally friendly solvolysis process for the chemical recycling of carbon fiber-reinforced epoxy resin polymers (CFREPs), utilizing methyl ethyl ketone (MEK) as a greener alternative to conventional acids such as nitric or acetic acid. The process employs a binary solvent system composed of MEK and hydrogen peroxide (H₂O₂), whose synergistic interaction enables the oxidative depolymerization of the epoxy matrix at a mild temperature of 70 °C. This approach not only achieves high resin decomposition rates but also facilitates the recovery of carbon fibers with minimal organic residue on their surfaces, making them suitable for reuse in recycled composite materials.

The process achieves effective resin decomposition across various solid-to-liquid ratios, with an optimal decomposition rate of 95% observed at a MEK/H₂O₂/H₂O molar ratio of 1/1/4. Characterization techniques such as scanning electron microscopy (SEM) and gel permeation chromatography (GPC) confirm that the recovered carbon fibers (rCFs) exhibit minimal resin residue, indicating substantial matrix degradation.

This recycling approach has proven successful for both cured and prepreg aeronautical composites, yielding rCFs that have been reintegrated into new thermoplastic and thermosetting matrix composites. Mechanical testing of these new composites compared to those made with virgin carbon fibers indicates that the recycling process does not degrade the structural integrity of the fibers. However, the process is nonselective, removing both the resin and the original sizing from the fibers. As a result, a resizing treatment is recommended to ensure optimal compatibility between the recycled fibers and the new matrix materials.

Glossary

Abbreviations

CFRP

carbon fiber-reinforced plastic

MEK

methyl ethyl ketone

H₂O₂

hydrogen peroxide

rCF

recycled carbon fibers

FTIR

Fourier transform infrared spectroscopy

GPC

gel permeation chromatography

TGA

thermogravimetric analysis

PA6

polyamide 6

CF

carbon fiber

RTM

resin transfer molding

rCFRP

recycled carbon fiber reinforced plastic

D _R

decomposition ratio

DMF

dimethylformamide

SEM

scanning electron microscopy

CFREP

carbon fiber-reinforced epoxy resin polymers

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.5c04208.

• TGA analysis of the initial materials, photographs of recovered carbon fibers from experiments, a diagram illustrating the generation of dioxiranes, synthetic schemes for both CFRP and CF prepreg epoxy resin, and the molecular weight distribution data of the decomposed epoxy resin recovered (PDF)

This study was funded by Environment of the Basque Government through the ELKARTEK 2023 Program.

The authors declare no competing financial interest.