Composites Science and Technology 82 (2013) 54–59

Contents lists available at SciVerse ScienceDirect

Composites Science and Technology

journal homepage: www.elsevier .com/ locate /compsci tech

Chemical recycling of carbon fibre/epoxy composites in a mixed solutionof peroxide hydrogen and N,N-dimethylformamide

0266-3538/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.compscitech.2013.04.002

⇑ Corresponding author. Tel.: +86 574 86685256; fax: +86 574 86685186.E-mail address: lijuan@nimte.ac.cn (J. Li).

Pinglai Xu, Juan Li ⇑, Jiangping DingNingbo Key Laboratory of Polymer Materials, Ningbo Institute of Material Technology and Engineering, Chinese Academy of Sciences, Ningbo, Zhejiang 315201, PR China

a r t i c l e i n f o a b s t r a c t

Article history:Received 16 March 2012Received in revised form 12 March 2013Accepted 2 April 2013Available online 18 April 2013

Keywords:A. RecyclingB. Carbon fibresC. Polymer-matrix composites(PMCs)

A two-step method to recycle high-quality carbon fibres (CFs) from carbon fibre/epoxy (CF/EP) compos-ites in high yield under mild conditions was reported in this paper. Firstly, the composites were pre-treated in acetic acid to be expanded and be layered to get larger surface area. Secondly, a synergisticoxidative degradation system is supposed to recover CF, which is a mixed solution of peroxide hydrogen(H2O2) and N,N-dimethylformamide (DMF) in a hermetic reactor. The structure and properties of recov-ered CFs were investigated. The results showed that clean CFs can be successfully recycled after the pre-treated composites was treated at 90 �C for 30 min in a solution of H2O2/DMF (1:1, v/v). Thedecomposition ratio (Dr) of EP in composites was more than 90%. The surface of the CFs was smooth withfew residues of EP observed by scanning electron microscope (SEM). The degree of graphitization ofrecovered CFs was decreased slightly tested by using Raman and X-ray diffraction (XRD) spectrum.The tensile strength of the recovered CFs was more than 95% of the virgin ones’ according to the singlefibre tensile test.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Carbon fibre/epoxy (CF/EP) composites have obtained more andmore attention because of their low density, high strength andmodulus, good corrosion resistance and desirable comprehensiveproperties etc. [1]. With the developing of technology, increasingcomposites are manufactured for application in high-tech fields.At the same time, as the volume of CF/EP usage grows, moreend-of-life products are generated as well as rejects and off-cuts[2,3]. CF/EP composites as one kind of thermoset resin matrix com-posites are insoluble and infusible, so it is difficult to be recycled byusing usual methods. Without an effective process, the waste of CF/EP composites will bring great pressure on the environment. Tosolve the problem, it is necessary to turn CF/EP composites into avaluable resource and close the loop in the composites life-cycle.In the past decades, the waste of CF/EP composites was mainly dis-posed in landfill or by combustion for little commercial profit [2].Now many countries, especially the European countries [4], putmany strict rules and laws on the disposal of the thermoset resinsinto effect. An efficient recycling route is highly needed for recy-cling the CFs with high value and reducing the pressure producedby the thermoset based composites on the environment.

Since the early 90s of 20th century, studies on recycling of fibrereinforced resin composites focused on four methods [5]:

mechanical recycling [6–9], pyrolysis process [10–12], fluidizedbed process [13–17] and chemical recycling [2,18–35].

Mechanical recycling as an original mean produces powderedfillers and fibre products. Kouparitsas et al. [8] and Palmer et al.[9] grinded the composites to achieve fillers or/and fibres ofwide-ranged lengths. By this process, the composites had beenrecycled, although long and high valuable fibres could not beobtained.

Pyrolysis process, as a common thermal process, gives chemicalproducts, fibres and fillers. Torres et al. [11] recycled the compos-ites by pyrolysing over 400 �C and obtained a complex liquid mix-ture of C5–C20 organic compounds that could be used as fuel oils.To some extent, at a high temperature, pyrolysis process givessome liquid products fuel, some useful solid residues and long fi-bres with high modulus, however, also releases a gas mixture ofCO and CO2 and so on.

Fluidized bed process as another thermal process can produceclean fibres and fillers with energy recovery. Pickering et al. [13–16] successfully recycled glass fibres by fluidized-bed process fromcomposites at a bed temperature of 450 �C and a fluidizing velocityof 1.3 m/s. And later Zheng et al. [17] recycled glass fibres from theprinted circuit boards at the temperature ranging from 400 �C to600 �C. The fluidized bed process really gives clean fibres and otherproducts but the reclaimed fibres would be highly oxidized, caus-ing property reduction.

Chemical recycling includes liquid-phase cracking, chemolysis,supercritical fluid process etc. Compared to the mechanical and

Table 1Dr of different samples under different reaction conditions.

No. DMF (ml) H2O2 (ml) Temperature (�C) Time (min)

P. Xu et al. / Composites Science and Technology 82 (2013) 54–59 55

thermal processes mentioned above, chemical recycling is noteffective to recycle the composites because not all the solution isfeasible to all composites. Developing method for chemical recy-cling of thermoset composites to useful CFs and organic com-pounds is also an important issue. The supercritical fluids whichhave the combined characteristics of liquid-like density, dissolvingpower, and gas-like viscosity and diffusivity have been used to re-cycle the composites. Several types of fluids were studied, forexample water, methanol, propanol etc. Bai et al. [2] recycled CFsin supercritical water with oxygen at 30 ± 1 MPa and 440 ± 10 �C.Liu et al. [20,21] and Piñero-Hernanz et al. [23,24] also studiedthe recycling of composites in near-critical or/and supercriticalwater. They found that resin removal was much higher using alkalicatalyst than that without any catalysts. Alcohol is also used whosecritical temperature is much lower than water. The recycled CFform alcohol has good properties.

Though many chemical solvents have been used to recycle theCFs from composites and some progress has been achieved, thereare few reports which focused on the oxidized degradation ofepoxy in hydrogen peroxide (H2O2). As a matter of fact the waterunder subcritical condition acts like strong oxidizer. From a‘‘green’’ chemistry perspective H2O2 is one of the best candidatesfor oxidants since water would be the only side product. It iswidely used in industry at low temperature and pressure and itseffect is similar to the subcritical water at high temperature andpressure. Based on this background, a mixed solution includingH2O2 and N,N-dimethylformamide (DMF) is supposed to decom-pose the CF/EP composites in this paper. In this system, the H2O2

is used as an oxidant to decompose the CF/EP composites whileDMF is used to dissolve the decomposed products and acceleratethe reaction of decomposition. The CFs may be recycled from theCF/EP composites under low temperatures and low pressure. Ther-mal gravimetric analysis (TGA) were used to measure the residuesof EP resins on the surface of the recovered CFs. Scanning electronmicroscope (SEM), Raman and X-ray diffraction (XRD) spectrumwere used to characterize the microstructure and element compo-nent of the recovered CFs. XQ-1/XD-1 single fibre tester wasadopted to measure the tensile strength of the recovered CFs.

1# 15 15 80 302# 15 15 85 303# 15 15 90 304# 15 15 100 305# 15 15 120 306# 15 15 135 307# 15 15 150 308# 15 15 100 59# 15 15 100 1010# 15 15 100 2011# 15 15 100 6012# 15 15 100 9013# 15 15 100 12014# 15 0 100 3015# 0 15 100 3016# 1.5 15 100 3017# 3 15 100 3018# 5 15 100 3019# 7.5 15 100 3020# 10 15 100 3021# 15 10 100 3022# 15 7.5 100 30

2. Experimental part

2.1. Materials

The composites were manufactured in our lab using a vacuumassisted resin infusion (VARI) process. The resin used in our exper-iment is JR-236 EP with the epoxide value of 0.54–0.57, and curingagent is acyclic amine JH-239 obtained from Changshu Jiafa chem-ical Co., Ltd., China. The commercially available non-crimp fabricwas provided by Changzhou Hongfa Zongheng Advanced MaterialsTechnology Co. Ltd., China. The used CF is T-700 12 K from JapanToray with a diameter of about 7 lm. The resin and curing agentwere mixed at a 100:30 weight ratio and then heated and degassedfor 15 min. The above mixture was soaked through the CF layersand cured at 70 �C for 2 h, and then 110 �C for 2 h [36]. The CF/EP composites were cut into slices of ca. 10 � 10 � 2.5 mm3 beforepretreatment. The acetic acid (G.R. grade), 30% H2O2 solution (A.Rgrade) and DMF (A.R. grade) were purchased from Sinopharmchemical reagent company and used without further purification.

Table 2Raman results of CFs treated under different reaction conditions.

Samples ID (cm�1) IG (cm�1) ID/IG

VF 1368.4 1583.3 1.853# 1364.6 1579.3 2.084# 1359.9 1574.9 2.215# 1361.4 1577.4 2.41

2.2. Decomposition of CF/EP composites

The composites sheets were put into a three-necked flask with25 ml acetic acid at refluxing temperature of 120 �C for 30 min. Thetreated composites were washed with acetone for several timesthen dried in vacuum at 100 �C for 12 h.

About 0.2 g pretreated CF/EP composites, 0–15 ml H2O2 and 0–15 ml DMF were put into a stainless100 ml autoclave without stir-ring. The hermetic autoclave was heated at 80–150 �C for 5–120 min. The detailed conditions are listed in Table 1. After thereaction, the recovered CFs were taken out of the autoclave, rinsedwith acetone for several times and then dried in vacuum at 100 �Cfor 24 h.

2.3. Characterization

2.3.1. Thermal gravimetric analysis (TGA)TGA experiments were performed on a Mettler Toledo TGA/

DSC1 Analyzer. About 5 mg specimens were heated from 25 �C to800 �C at a heating rate of 10 �C/min under N2 atmosphere(50 ml/min).

2.3.2. Decomposition ratio (Dr)Approximately 20 mg composites samples were heated from

room temperature to 500 �C and kept at 500 �C for 20 min in themuffle furnace. Then the products were weighed by electron bal-ance after cooling to the room temperature. The Dr of the compos-ites was calculated according to the amount of solid compositionafter treatment using the following formula:

Dr ¼W1 �W2

W0ð1Þ

where W1 represents the mass of composites before decomposition,W2 represents the mass of solid residues after decomposition, andW0 represents the mass of epoxy resin in the composites beforedecomposition.

2.3.3. Scanning electron microscopy (SEM)The samples were sputter-coated with a conductive layer of

platinum, and then their surface morphologies were observed by

56 P. Xu et al. / Composites Science and Technology 82 (2013) 54–59

a Hitachi S-4800 scanning electron microscopy with an 8 kV accel-erated voltage.

2.3.4. Raman spectroscopyRaman spectra were recorded on a Renishaw RM1000 Raman

spectrograph. The laser power measured at the samples was below10.0 mW for 514 nm radiation. The scanning wavenumber rangewas 100 cm�1 to 3200 cm�1.

2.3.5. X-ray diffraction (XRD)X-ray diffraction measurements were made from the filament

of virgin and recovered CFs. The diffraction studies were performedin reflection mode using X-ray diffractometer, Bruker AXS D8 Dis-cover using Cu Ka radiation at a scan rate of 0.2�/s in a 2h of 5–60�,and operated at 30 kV and 20 mA.

2.3.6. Single fibre tensile testThe single fibre tensile property of the recovered CF was tested

by Donghua university XQ-1/XD-1 single fibre tester. Because therecovered CF was short and disordered, the samples cannot betested according to the standard method. The strength of CF wasdetermined as following. Single filament was selected from T70012 k with a length about 2.5 cm. The distance between two clampswas 2 cm. The stretching speed was 20 mm/min. At least 10 fila-ments were tested for each sample.

3. Results and discussion

3.1. The EP resin residuals on the surface of the recovered CFs

A two-step method of chemical recycling of EP/CF composites isused. The processing is shown in Scheme 1. Firstly, the compositeswere pretreated in acetic acid at 120 �C for 30 min to be expandedand layered to get larger surface area which benefits to the nextimmersion of solvent in composites so as to accelerate the reactionof degradation. Usually the interaction between layers of compos-ites is poorer than that in layers of fabrics, the breakage would oc-cur between layers when the composites are damaged. Moreover,the mass remains the same before and after pretreatment becauseof no chemical reaction, thus the solvent can be reused and theafter-treatment processing is simplified. Secondly, a synergistic

Scheme 1. Road of the decompo

70 80 90 100 110 120 130 140 150 1600

20

40

60

80

100

Dr

(%)

Temperature (OC)

0 20 400

20

40

60

80

100

Dr (

%)

Tim

Fig. 1. Relationship between Dr and (a) tem

oxidative degradation system is assumed, which is a mixed solu-tion of H2O2 and DMF. The H2O2 systems generate hydroxyl radi-cals (�OH) which are powerful oxidizing species. �OH can oxidizeorganic compounds (RH) producing organic radicals (�R), whichare highly reactive and can be further oxidized. In the other hand,the de-crosslinking components are dissolved in DMF and the reac-tion is promoted. The CF/EP composites were treated in differenttemperatures, time and volume ratios of DMF and H2O2 solution.Finally, clean CFs and degraded products in deep yellow liquidwere obtained. In our experiment, to study the decomposition ratioaccurately, excessive solvents were used. The treated condition ismild so it is promising to be scaled up in the future.

To study the decomposition of EP in composites, the CFs resi-dues before and after treatment are weighed exactly and the Dr

is calculated according to Formula 1. The results are shown inFig. 1. Keeping reaction time (30 min) and DMF/H2O2 (1:1, v/v) ra-tio unchanged, the Dr increases apparently with the increasing ofreaction temperature as shown in Fig. 1a. The Dr is below 60%when the temperature is lower than 90 �C, but it achieves 93.4%after treated at 90 �C for 30 min. The Dr increases slightly whentemperature is above 90 �C. In the one hand, enough energy is pro-vided to the system at a high temperature, thus the decompositionreaction could develop smoothly. In the other hand, because thereacting composite is solid, the mass transfer plays an importantrole in this system. The energy is enough above 90 �C, however,the radicals produced by H2O2 cannot attack the inside EP easily.Therefore the decomposition reaction happens on the surface ofEP and develops gradually.

Reaction time is another factor for the decomposition process.From Fig. 1b, it could be found that the Dr is only 2% after reactingat 100 �C for 5 min. Prolonging the time to 10 min, the Dr can reach75.9%. When the reaction time is over 30 min, the Dr is more than90% and changed slightly. This phenomenon is similar to the re-sults obtained under different temperatures. It needs time to finishthe initiation, propagation of radicals, diffusion and the oxidativedecomposition of EP resins. Furthermore, it also takes some timeto decompose the EP on CFs layer by layer. Enough reaction time(P30 min) can guarantee over 90% EP resins decomposed.

Volume ratio of DMF and H2O2 solution (VDMF/VH2O2) also hassignificant influence on the Dr of EP. According to the results shownin Fig. 1c, single H2O2 or DMF can result in a Dr of 55.6 (15#) and23.0 (14#), respectively. The Dr is increased to 92.6% for the mixed

sition of CF/EP composites.

60 80 100 120e (min)

0

20

40

60

80

100

22#21#20#19#18#17#16#15#14#

Dr (

%)

Samples

perature, (b) time and (c) VDMF/VH2O2.

100 200 300 400 500 600 700 800

70

75

80

85

90

95

100

a b c d e

Wei

ght/%

Temperature /

Fig. 2. TGA of (a) CF/EP composites and CFs obtained in (b) H2O2 solution, (c) DMFand (d and e) DMF/H2O2 mixed solution at 90 �C and 150 �C for 30 min respectively.

P. Xu et al. / Composites Science and Technology 82 (2013) 54–59 57

solution system (4#) with the same temperature and time. In addi-tion, incorporating a little H2O2 into DMF or the opposite, the Dr

(16–21#) can be increased sharply. Excessive DMF might dilutethe concentration of H2O2 and obstruct the reaction. The Dr is high-er than 90% when the ratio of VDMF/VH2O2 is between 0.1 and 1.5.Therefore, appropriate VDMF/VH2O2 is helpful for the mass transferand the decomposition of EP resins. The Dr achieves 99.1% for thecomposites decomposed in solution (VDMF/VH2O2 = 1:2), indicatingan optimized effect between H2O2 and DMF on the decompositionof EP resin.

Fig. 2 is the TGA results of CF/EP composites and recovered CFsobtained in different conditions. Because the CFs does not decom-pose under N2 atmosphere, the weight loss means amount of resinresidues on the surface of the recovered CFs. The weight of samplesdecreases gradually with the increasing of temperature, so theweight at 800 �C is chose to calculate the Dr. According to Fig. 2a,the EP resin content in initial composite sample was 30.0 wt.% at

Fig. 3. SEM micrograph of (a) CF/EP composites, (b) virgin CFs, (c and d) recovered CF

800 �C. The EP residues decreases to 25.0 wt.% after being treatedin DMF, indicating a 16.7% of Dr determined from Formula 1.Though DMF is not a good reactive solvent, part of EP could bedecomposed. When treated in H2O2 solution, the Dr of EP is51.3%. The oxidation of H2O2 is positive for the decomposition ofEP. Moreover, the thermal degradation of the sample begins at50 �C observed in Fig. 2b, which means some de-crosslinking ofthe resin. The weight reaches to 98.5 wt.% for the solid residuesrecovered from a mixed solution of DMF/H2O2 (1/1) at 90 �C for30 min as shown in Fig. 2d. It is suggested that the EP resins hasbeen decomposed 95.2%. The synergic effect is obvious betweenDMF and H2O2 in the decomposition of EP resins which is agree-ment with the results by weighing. Of course, the Dr increased to99% with the reaction temperature increasing to 150 �C. Becausethe decomposition of composites is a reaction from solid to liquid,it is difficult to make the reaction performing uniformly. WhileTGA curve is obtained by extracting only 5 mg specimens, it is onlya fraction of the recovered CFs. Therefore calculating Dr from TGA isnot an accurate method, and it can be used as a kind of qualitativeanalysis. On the whole, the trend of TGA results is in accordancewith that obtained by weighing. The structure and properties ofthe recovered CFs are characterized as following.

3.2. Surface morphology of the recovered CFs

Surface morphology of the recovered CFs was examined bySEM. The CF is adhered by EP resin in the composites as shownin Fig. 3a. After being treated, the surface of the recovered CFs isclean and smooth, no grooves or holes are found. Compared tothe virgin CFs, some white spots are observed on the surface ofrecovered CFs (Fig. 3c and d) which should be the residues of EP re-sin. The diameter of the virgin CF is about 7 lm, the recovered CFsobtained in different conditions changes slightly observed by SEM.The surface micrograph of the recovered CFs suggests that the CF isnot damaged dramatically in the processing.

s obtained from the samples treated at 90 �C and 150 �C for 30 min respectively.

Table 3XRD results of CFs treated under different reaction conditions.

Samples d002 (nm) 2h(002) (�) FWHM (�) Lc (nm) Lc/d002

VF 0.3520 25.28 1.002 8.12 23.13# 0.3519 25.29 1.008 8.07 22.94# 0.3532 25.29 1.025 7.94 22.57# 0.3530 25.21 1.039 7.83 22.2

58 P. Xu et al. / Composites Science and Technology 82 (2013) 54–59

3.3. Structure and properties of the recovered CFs

As know to all, the structure connects tightly with the proper-ties of CFs, higher degree of graphitization means higher strength.Raman spectrogram combines a prominent surface selectivity withan exceptional sensitivity to the degree of structural disorder [37].Carbon mainly shows two peaks in the first order (1000–2000 cm�1). One near 1580 cm�1 is corresponding to an ideal gra-phitic lattice called G band while the other near 1360 cm�1 is thecharacteristic peak of sp3 state of C called D band which is due tothe existence of structural disorder [38]. The D to G band inte-grated intensity ratio (ID/IG) is thus a parameter to quantify the de-gree of disorder. Usually, the smaller value of ID/IG indicates thehigher degree of graphitization of the CFs. The Raman results ofthe virgin and recovered CFs are shown in Fig. 4 and Table 2. Itshows that the value of ID/IG of 3# CFs is 2.08 which is higherslightly than that of the virgin CFs (1.85). This tells us that the de-gree of graphitization of the recovered CFs decreases slightly. Withthe increasing of the temperature, the values of ID/IG of 4# and 5#increase suggesting more damage of CFs.

XRD spectra were also used to test the structure of CFs as shownin Fig. 5. The detailed data are listed in Table 3. It is seen that astrong peak ((002) plane) exists at ca. 2h = 25.3�. The interlayerspacing (d002) and apparent crystallite thickness (Lc) derived from(002) reflection are also used as an experimental parameter to as-sess the structural properties of the CFs [38,39]. Layers of thestacked crystallite are used to measure the degree of graphitizationof CF determined by Lc/d002. It is found that the layers of the

800 1000 1200 1400 1600 1800 2000 2200

GD

5#

4#

3#

Inte

nsity

(a.

u.)

/cm-1

VF

Fig. 4. Raman spectra of the CFs recovered at different temperature.

10 20 30 40 50 60

7

4#

3#

VF

002

Inte

nsity

(a.

u.)

2

Fig. 5. XRD spectra of the CFs recovered at different temperature.

stacked crystallite of the recovered CFs (22.7) are comparable withthe virgin ones’ (23.1). It reveals that the recovered CFs retains highdegree of graphitization of the virgin CFs. These results indicatethat the structure of CFs is damaged slightly during the recyclingprocess. Combining the SEM, Raman and XRD results, it is believedthat the structure of CFs is not damaged dramatically, and this canbe testified by the tensile strength discussed below.

The tensile strength of single fibre is shown in Fig. 6. The tensilestrength of virgin CFs is 2.81 GPa which is lower that the value pro-vided by Toray because the testing is not standard. The 4# CFswhich was recovered at 100 �C for 30 min has a tensile strengthof 2.76 GPa, about 97% of the virgin one’s. The tensile strength de-creases with the treated temperature increasing. After being trea-ted at 135 �C for 30 min, the tensile strength is 2.43 GPa which isdecreased 17% compared to the virgin ones’. The tensile strengthof recovered CFs remains higher than 90% of the virgin ones’ ifthe temperature is lower than 120 �C. The results of the single ten-sile strength are in accordance with the structure characterizationshowed above. The results reveal that the structure and propertiesof the recovered CFs are not destroyed dramatically.

Based on the results, the decomposition process of CF/EPcomposites in DMF and H2O2 is considered to be divided into fiveparts: initiation, diffusion, oxidative decomposition, dissolutionand separation, as shown in Scheme 2. Firstly, the oxidant H2O2

will produce radicals under selected conditions. Because thedecomposition reaction occurs between solid and liquid, it is dif-ferent from a homogeneous reaction. The mass transfer plays akey role in the system. Only the active radicals are diffused tothe surface of composites, the propagation and transfer of radicalscan be induced. Then the EP will be decomposed by oxidation. En-ough concentration of H2O2 ensures the decomposition of EP resinsproceeding entirely. Due to the solid prevents the reactive compo-nent from contacting with the inside EP, the decomposition can de-velop gradually. DMF is a good reagent for decomposed products ofEP, which will be dissolved and removed easily. Thus the inside EPcan be attacked and decomposed. Therefore the reaction will beaccelerated by the introducing of DMF. The time (P30 min), tem-perature (P90 �C) and appropriate VDMF/VH2O2 (0.1–1.5) ratio isnecessary for the decomposition of over 90% EP resins. Though it

0

1

2

3

4

5

2.43 0.642.58 0.712.76 0.52

VF 6#5#

Tens

ile

stre

ngth

(G

Pa)

4#

2.81 0.81

Fig. 6. Tensile strength of single fibre of the CFs recovered at different temperature.

Scheme 2. Decomposition process of CF/EP composites in DMF and H2O2.

P. Xu et al. / Composites Science and Technology 82 (2013) 54–59 59

is not a flash reaction, the condition is mild and feasible. It is a pro-gress for the chemical recycling of CF/EP composites. Leaving cleanCF with high quality is important for its reusing. The decomposedproducts and mechanism of EP is complex and will be studied inour following work.

4. Conclusions

A two-step method of chemical recycling of CF/EP composites inmild conditions was studied. The composites were first treated inacetic acid to be expanded and layered, and then were decomposedin a mixed solution with DMF and H2O2. The CFs can be taken outfrom the CF/EP composites at 90 �C for only 30 min with VDMF/VH2O2 = 1:1. The decomposition ratio of EP is above 90%, and thestrength of CF remains more than 95% of its original strength.The SEM reveals that the surface of recovered CFs is smooth andfew EP resins are adhered, suggesting the decomposition of themajor mass of EP. The degree of graphitization of recovered CFsis decreased slightly compared to the virgin ones’. The single ten-sile strength of the CFs recovered at 100 �C for 30 min is2.76 GPa, which is about 98% of the virgin ones’. In a word, theCFs with high quality can be recovered from CF/EP composites inmild conditions. It is promising to be scaled up in the future.

Acknowledgements

This work was supported by the Director Fund of Ningbo Insti-tute of Material Technology and Engineering, Chinese Academy ofSciences, and Program for Ningbo Innovative Research Team (Grant2009B21008).

References

[1] Sun ML, editor. Principle and technology of epoxyapplications. Beijing: Machinery Industry Press; 2002.

[2] Bai YP, Wang Z, Feng LQ. Chemical recycling of carbon fibres reinforced epoxyresin composites in oxygen in supercritical water. Mater Des2010;31:999–1002.

[3] Pimenta S, Pinho ST. Recycling carbon fibre reinforced polymers for structuralapplications: technology review and market outlook. Waste Manage2011;31:378–92.

[4] Robert T. Rapid growth forecast for carbon fibre market. J Reinf Plast Compos2007;52:3–10.

[5] Pickering SJ. Recycling technologies for thermoset composite materials –current status. Composites Part A 2006;37:1206–15.

[6] Ralph BJ, W. DG. Recycling SMC. In: The 46th annual conference, compositeinsititute, the society of the plastics industry, Inc.; 1991. p. 1–6.

[7] Bledzdi AK. The use of recycled fibre composites as reinforcement forthermosets. Mekhanika Kompozitnykh Materialov 1993;29:473–7.

[8] Kouparitsas CE, Kartalis CN, Varelidis PC, Tsenoglou CJ, Papaspyrides CD.Recycling of the fibrous fraction of reinforced thermoset composites. PolymCompos 2002;23:682–9.

[9] Palmer J, Ghita OR, Savage L, Evans KE. Successful closed-loop recycling ofthermoset composites. Composites Part A 2009;40:490–8.

[10] Torres A, de Marco I, Caballero BM, Laresgoiti MF, Legarreta JA, Cabrero MA,et al. Recycling by pyrolysis of thermoset composites: characteristics of theliquid and gaseous fuels obtained. Fuel 2000;79:897–902.

[11] Meyer LO, Schulte K, Grove-Nielsen E. CFRP-recycling following a pyrolysisroute: process optimization and potentials. J Compos Mater 2009;43:1121–32.

[12] Jiang G, Pickering SJ, Walker GS, Bowering N, Wong KH, Rudd CD. Softionisation analysis of evolved gas for oxidative decomposition of an epoxyresin/carbon fibre composite. Thermochim Acta 2007;454:109–15.

[13] Kennerley JR, Fenwick NJ, Pickering SJ, Rudd CD. The properties of glass fibresrecycled from the thermal processing of scrap thermoset composites. J VinylAddit Techn 1997;3:58–63.

[14] Pickering SJ, Kelly RM, Kennerley JR, Rudd CD, Fenwick NJ. A fluidised-bedprocess for the recovery of glass fibres from scrap thermoset composites.Compos Sci Technol 2000;60:509–23.

[15] Yip HLH, Pickering SJ, Rudd CD. Characterisation of carbon fibres recycled fromscrap composites using fluidised bed process. Plast Rub Compos2002;31:278–82.

[16] Jiang G, Pickering SJ, Walker GS, Wong KH, Rudd CD. Surface characterisationof carbon fibre recycled using fluidised bed. Appl Surf Sci 2008;254:2588–93.

[17] Zheng YH, Shen ZG, Ma SL, Cai CJ, Zhao XH, Xing YS. A novel approach torecycling of glass fibres from nonmetal materials of waste printed circuitboards. J Hazard Mater 2009;170:978–82.

[18] Yamada K, Tomonaga F, Kamimura A. Improved preparation of recycledpolymers in chemical recycling of fibre-reinforced plastics and molding of testproduct using recycled polymers. J Mater Cycles Waste Manage2010;12:271–4.

[19] Kamimura A, Akinari Y, Watanabe T, Yamada K, Tomonaga F. Efficient chemicalrecycling of waste fibre-reinforced plastics: use of reduced amounts ofdimethylaminopyridine and activated charcoal for the purification ofrecovered monomer. J Mater Cycles Waste Manage 2010;12:93–7.

[20] Liu YY, Shan GH, Meng LH. Recycling of carbon fibre reinforced compositesusing water in subcritical conditions. Mater Sci Eng A 2009;520:179–83.

[21] Liu YY, Meng LH, Huang YD, Du JJ. Recycling of carbon/epoxy composites. JAppl Polym Sci 2004;95:1912–6.

[22] Goto M. Chemical recycling of plastics using sub- and supercritical fluids. JSupercrit Fluid 2009;47:500–7.

[23] Pinero-Hernanz R, Garcia-Serna J, Dodds C, Hyde J, Poliakoff M, Cocero MJ,et al. Chemical recycling of carbon fibre composites using alcohols undersubcritical and supercritical conditions. J Supercrit Fluid 2008;46:83–92.

[24] Pinero-Hernanz R, Dodds C, Hyde J, Garcia-Serna J, Poliakoff M, Lester E, et al.Chemical recycling of carbon fibre reinforced composites in near critical andsupercritical water. Composites Part A 2008;39:454–61.

[25] Suyama K, Kubota M, Shirai M, Yoshida H. Degradation of crosslinked unsaturatedpolyesters in sub-critical water. Polym Degrad Stab 2007;92:317–22.

[26] Mormann W, Frank P. (Supercritical) ammonia for recycling of thermosetpolymers. Macromol Symp 2006;242:165–73.

[27] El Gersifi K, Durand G, Tersac G. Solvolysis of bisphenol A diglycidyl ether/anhydride model networks. Polym Degrad Stab 2006;91:690–702.

[28] Dang WR, Kubouchi M, Sembokuya H, Tsuda K. Chemical recycling of glassfibre reinforced epoxy resin cured with amine using nitric acid. Polymer2005;46:1905–12.

[29] Sato Y, Kondo Y, Tsujita K, Kawai N. Degradation behaviour and recovery ofbisphenol-A from epoxy resin and polycarbonate resin by liquid-phasechemical recycling. Polym Degrad Stab 2005;89:317–26.

[30] Kojima M, Tosaka M, Ikeda Y. Chemical recycling of sulfur-cured naturalrubber using supercritical carbon dioxide. Green Chem 2004;6:84–9.

[31] Dang WR, Kubouchi M, Yamamoto S, Sembokuya H, Tsuda K. An approach tochemical recycling of epoxy resin cured with amine using nitric acid. Polymer2002;43:2953–8.

[32] Xiu FR, Zhang FS. Materials recovery from waste printed circuit boards bysupercritical methanol. J Hazard Mater 2010;178:628–34.

[33] Jiang G, Pickering SJ, Lester EH, Turner TA, Wong KH, Warrior NA.Characterisation of carbon fibres recycled from carbon fibre/epoxy resincomposites using supercritical n-propanol. Compos Sci Technol2009;69:192–8.

[34] Jiang GZ, Pickering SJ, Lester EH, Warrior NA. Decomposition of epoxy resin insupercritical isopropanol. Ind Eng Chem Res 2010;49:4535–41.

[35] Hyde JR, Lester E, Kingman S, Pickering S, Wong KH. Supercritical propanol, apossible route to composite carbon fibre recovery: a viability study.Composites Part A 2006;37:2171–5.

[36] Ding JP, Pan LJ, Fan XY, Tan YF. Study on the mechanical properties of domesticCCF300 darbon fibre four axial directions non-crimp fabric laminates. Hi-TechFibre Appl 2010;35:26–31.

[37] Montes-Moran MA, Young RJ. Raman spectroscopy study of HM carbon fibres:effect of plasma treatment on the interfacial properties of single fibre/epoxycomposites – Part I: Fibre characterisation. Carbon 2002;40:845–55.

[38] Li DF, Wang HJ, Wang XK. Effect of microstructure on the modulus of PAN-based carbon fibres during high temperature treatment and hot stretchinggraphitization. J Mater Sci 2007;42:4642–9.

[39] Li DF, Wang HJ, He F, Wang XK. Structure and properties of T300 and T700carbon fibres. New Carbon Mater 2007;22:1–6.

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具