Reclamation of carbon fibers and added-value gases in a pyrolysisbased composites recycling process

Reclamation of carbon fibers and added-value gases in a pyrolysis-based composites recycling process A. Lopez-Urionabarrenechea1*, N. Gastelu1, E. Acha1, B.M. Caballero1, A. Orue2, A. JiménezSuárez3, S.G. Prolongo3, I. de Marco1 1 Chemical and Environmental Department, Faculty of Engineering of Bilbao. University of the Basque Country (UPV/EHU). Plaza Ingeniero Torres Quevedo 1, 48013 Bilbao (Spain). 2 Chemical and Environmental Department, Faculty of Engineering of Gipuzkoa. University of the Basque Country (UPV/EHU). Plaza Europa 1, 20018 Donostia-San Sebastian (Spain). 3 Materials Science and Engineering Area, ESCET. University Rey Juan Carlos. C/Tulipán s/n, Móstoles, Madrid (Spain) Corresponding author information: This is the accepted manuscript of the article that appeared in final form in Journal of Cleaner Production 273 : (2020) // Article ID 123173, which has been published in final form at https://doi.org/10.1016/ j.jclepro.2020.123173. © 2020 Elsevier under CC BY-NC-ND license (http://creativecommons.org/licenses/bync-nd/4.0/) 1 Abstract 1 Carbon fibers reclamation from waste composites by pyrolysis is developing fast. Pyrolysis 2 allows the reclamation of the fibers by thermal decomposition of the polymeric resin, but the 3 material value of these resins is not yet recovered. In this research work, the possibility of 4 obtaining high-value gaseous fractions by treating the vapors generated in the decomposition 5 of a poly(benzoxazine) containing carbon fiber waste composite has been studied. The 6 experiments have been carried out in a lab-scale facility consisting of two reactors in series. In 7 the first reactor, the pyrolysis of the waste is carried out at 500 °C, while in the second reactor 8 the vapors coming from the first reactor are treated at 900 °C. Such treatment enables to triple 9 the amount of gas in comparison to conventional pyrolysis. Besides, this gas contains 56% by 10 volume of H2, which means the recovery of approximately 60% of the hydrogen contained in the 11 resin. On the other hand, clean carbon fibers with mechanical properties within commercial 12 values have been also obtained. 13 14 Keywords: carbon fiber, pyrolysis, carbon composites, recycling, poly(benzoxazine), secondary 15 raw materials. 16 17 Wordcount: 7915 18 19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 2 1. Introduction 20 21 Carbon fiber (CF) is a very low-density material with excellent mechanical, electrical and thermal 22 properties, which is almost exclusively used as reinforcement of plastic matrix composite 23 materials, called carbon fiber reinforced plastics/polymers (CFRP) (Giorgini et al., 2015). Behind 24 the generic definition of CFRP, there is a wide variety of products covering large ranges of 25 physical-chemical properties. This is because such properties depend fundamentally on the 26 quality of the fibers - which can be manufactured almost to measure according to the needs in 27 service - and the quantity, length and orientation of the fibers within the plastic matrix - both 28 length and alignment increasing the reinforcement capacity of the fibers - (Newcomb, 2016; 29 William D. Callister Jr., 2014). This circumstance gives CFRP a versatility that explains their 30 massive and growing industrial and urban use (Holmes, 2017a; Liu et al., 2019; Vo Dong et al., 31 2019). 32 33 However, CF production is a process with high environmental and energy impact. 96% of CF is 34 produced from poly(acrylonitrile) (PAN) fibers (Newcomb, 2016). PAN is obtained by 35 polymerization of acrylonitrile, which in turn is the product of the reaction between propylene 36 and ammonia, both of them originally coming from oil (Naqvi et al., 2018). PAN fibers are 37 carbonized in a temperature range between 1000 °C to 3000 °C, depending on the specific 38 properties desired for the CF. Finally, CF are conditioned through surface treatments (Naqvi et 39 al., 2018; Newcomb, 2016). In summary, CF is a fossil product that is manufactured through a 40 multi-stage process where very high temperatures, solvents, finishing oils and surface treatment 41 resins are required. Additionally, atmospheric emissions of toxic compounds such as HCN, NH3 42 or NOx from furnaces and reactors are produced (Lefeuvre et al., 2017; Vo Dong et al., 2018). 43 44 One way to reduce the impact of the growing CF production is to recover them from residual 45 CFRP and re-manufacture composite materials with the reclaimed fibers (Hermansson et al., 46 2019). There is still much to be developed in this field, as although several methods of CF 47 recovery have been investigated, landfill and incineration are today the two most common ways 48 of managing CFRP waste (Liu et al., 2017; Vo Dong et al., 2018; Yang et al., 2012). In order to 49 recover the CF from a composite material it is necessary to remove the polymeric resin where 50 they are embedded. This can be done by dissolution (chemical methods) and by decomposition 51 (thermal methods) (Pimenta and Pinho, 2011; Rybicka et al., 2016). Recently published 52 economic, environmental and human health impact assessments comparing these alternatives 53 point to the latter as the option with lowest impact, particularly when the thermal method is 54 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 3 pyrolysis (Khalil, 2018; Vo Dong et al., 2018). Consequently, pyrolysis is the most mature 55 technology, with the highest TRL and the largest number of industrial implementations, despite 56 the fact that chemicals are not recovered from the resin (it is burned) while dissolution methods 57 are theoretically capable of recovering the fibers and some monomers, mainly phenol and 58 derivatives (Kim et al., 2019; La Rosa et al., 2016). 59 60 Pyrolysis is heating in the absence of oxygen. In the case of CFRP waste, the final temperature is 61 usually between 400 and 600 °C, at which most resins are broken down but do not CF (Meng et 62 al., 2018; Meyer et al., 2009; Wong et al., 2010). The decomposition of the polymeric resin 63 produces gases, vapors (liquids at room temperature) and a carbonaceous product (char) that 64 forms a light layer covering the carbon fibers. This layer of char can prevent the proper adhesion 65 of the reclaimed fibers with new resins at the time of reuse, so it is necessary a post-treatment 66 in the presence of air to eliminate it (Onwudili et al., 2016). This post-treatment must be 67 controlled (not very high temperature and not very long time) in order to avoid the degradation 68 of the fibers by combustion (López et al., 2013; Nahil and Williams, 2011). According to 69 specialized literature, this process does not significantly damage the CF, which retain their 70 tensile modulus of elasticity and 90% of their tensile strength (Holmes, 2018; Pimenta and Pinho, 71 2014). 72 73 However, the recycling of CFRP waste by pyrolysis has not been launched definitively. The 74 studies indicate that it is still necessary to make the recycling process more profitable. There are 75 two areas with great potential for improvement: (1) obtaining longer fibers and developing 76 techniques for aligning short and long recycled carbon fibers (Holmes, 2018, 2017b; Meng et al., 77 2018) and (2) material recovery from polymeric resins (Cousins et al., 2019; Vo Dong et al., 2018). 78 The latter is not an easy task since CFRP pyrolysis liquids are complex mixtures of nitrogenous, 79 oxygenated and sulphur organic compounds, often mixed with water. On the other hand, the 80 gases, composed of carbon monoxide and dioxide, hydrogen and light hydrocarbons, are 81 generated in low quantity (López et al., 2013; Nahil and Williams, 2011; Yang et al., 2015). 82 83 The gases could be used as an alternative fuel, but the benefit obtained from them would not 84 compensate for the problem generated by the liquid product, whose industrial use is unlikely, 85 with the result that it would be classified as a hazardous waste. Therefore, it is evident that a 86 treatment is necessary to try to improve the composition of liquids and/or to eliminate them in 87 favor of the gaseous fraction. However, as far as we are concerned, there is no evidence from 88 any research group that has investigated this issue to date. The authors of this article have 89 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 4 developed a method based on the thermo-catalytic treatment of such gases and vapors before 90 condensation (Gastelu et al., 2019, 2018). This method avoids the emission of gaseous products 91 from resin combustion and recovers added-value gases. The objective of the present paper is to 92 determine the maximum performance of the treatment in the absence of catalyst, by studying 93 the influence of the filler used treatment reactor. 94 95 2. Materials and methods 96 97 2.1. CFRP waste 98 The CFRP waste are pre-preg pieces from aeronautical industry composed of a 99 poly(benzoxazine) type polymeric resin and Toray T300/3k carbon fibers fabric in the form of 100 plain weave. Table 1 shows some chemical properties of the CFRP sample. A deeper discussion 101 of these properties and the thermal behavior of this sample has been reported elsewhere 102 (Gastelu et al., 2018). 103 104 Table 1. Fiber content, proximate analysis and elemental analysis of the waste CFRP (wt.%). 105 Fiber content 61.0 Proximate analysis Moisture 0.5 Volatiles 28.4 Fixed carbon1 71.0 Ash 0.1 Elemental analysis C 86.2 H 2.2 N 4.8 S 1.3 Cl u.d.l.2 Others3 4.9 1 By difference 106 2 Under detection level (<0.1 wt.%) 107 3 Others = 100-(C+H+N+S+Moisture+Ash) 108 109 2.2. Lab-scale installation for carbon fiber reclamation 110 The experiments were carried out in a lab-scale installation consisting of two reactors connected 111 in series and a section for condensation of liquid products and cleaning of gases (see Figure 1). 112 The first reactor (3) is a non-stirred tank reactor 15 cm in diameter and 21 cm in height. The 113 second reactor (6) is a fixed bed tubular reactor 2.5 cm in diameter and 50 cm in length. The two 114 reactors are made of 316 stainless steel and are externally heated by electric furnaces. The 115 output of the second reactor (9) is connected to three condensers (11, 12, 13) and an activated 116 carbon column (Panreac Activated Carbon No. 1 QDP). The non-condensable products pass 117 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 5 through the activated carbon column (14) and then they are collected in gas sampling bags (16, 118 17). 119 120 121 122 1 Rotameter 11 Condenser 1 2 Pyrolysis / oxidation reactor 12 Condenser 2 3 Pyrolysis / oxidation furnace 13 Condenser 3 4 Furnace thermocouple 14 Activated carbon column 5 Reactor thermocouple 15 T valve 6 Tubular reactor 16 Gas bag 1 7 Tubular furnace 17 Gas bag 2 8 Tubular reactor thermocouple 18 Tubular furnace temperature-control 9 Vapor and gas outlet 19 Pyrolysis / oxidation furnace temperature-control 10 Solid bed material 20 Computer Figure 1. Flowsheet of the pyrolysis installation. 123 124 In a typical experiment, 100 g of CFRP waste (cut into 20 x 2 cm sheets) is introduced into the 125 first reactor, which is heated up at 3 °C min-1 to 500 °C for 30 minutes and is fed with a 1 L min126 1 flow rate of N2. The generated gases and vapors pass through the tubular reactor, heated at 127 900 °C, where a second thermal cracking takes place. This reactor is filled with a solid bed 128 material in order to improve fluid dynamics and avoid temperature gradients. Once the pyrolysis 129 stage is completed, the feed gas is changed to synthetic air in order to accomplish an oxidation 130 stage at 500 °C for 45 minutes, using an airflow rate of 1.5 L min-1. This paper presents the results 131 of five different experiments. The first one (E1) is a conventional pyrolysis test, without 132 treatment of gases and vapors. In the other four experiments (E2-E5) such treatment does occur 133 and the influence of the characteristics of the solid bed placed within the tubular reactor is 134 studied. 135 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 6 2.3. Materials employed in the solid bed 136 Three types of materials were used as fillers of the tubular reactor. On the one hand, silicon 137 carbide (CSi) provided by Navarrosic S.A., with two different particle size distributions, 1.5 mm 138 < x < 2 mm and 0.5 mm < x < 1 mm. CSi is a refractory material presenting good thermal 139 conductivity and high stability in the conditions of the vapors treatment (Dey et al., 2014). It is 140 widely used as solid bed in chemical reactors on a laboratory scale, since it improves fluid 141 dynamics and heat transport throughout them (Ortega et al., 2018; Song et al., 2018; Tillmann 142 et al., 2018). 143 144 The second solid bed material was made from defective bricks of composition 50-65% Al2O3, 30145 40% SiO2, 1-2% Fe2O3, 0.5-1.5% Na2O+K2O in weight, provided by Beroa S.L. Originally brick 146 shaped, they were ground to 1.5 < x < 2 mm particle size. This is a highly refractory material 147 (Seger Cone num. 37), so its function was expected to be equivalent to that of CSi but at a much 148 lower cost (they are waste bricks). In addition, the presence of alkali metal oxides and iron oxide 149 could have some catalytic effect that could be beneficial to the process. 150 151 Finally, a metallurgical coke of low volatile content provided by Befesa Zinc Aser S.A. was also 152 employed as solid bed material. Originally it was in the form of small briquettes, so it was ground 153 to 1.5 < x < 2 mm particle size. The objective of using a carbonaceous substance as solid bed is 154 to observe its catalytic action capacity in contact with CFRP decomposition substances, since 155 carbonaceous materials are widely tested substances for the treatment of oxygenated 156 substances and tars (Elkasabi et al., 2017; Ravenni et al., 2018). In this specific case, a 157 metallurgical coke has been selected because it is a “low-in-volatiles” material and therefore 158 relatively stable at the temperature of the treatment carried out at the tubular reactor (900 °C). 159 160 Table 2 shows the most representative textural parameters of the fillers. In general, they show 161 a very low surface area, so they can practically be considered non-porous, especially CSi. As 162 regards the other analyses carried out on these materials, the acidity analyses do not show any 163 appreciable acidic characteristics, so that neither coke nor refractory can be considered as an 164 acid material. In addition, the refractory does not exhibit reducible species with H2 (TPR) or 165 interaction with CO, so it can be considered a substance with stable metal oxides. The data 166 obtained in these analyses are not presented in the manuscript because they lack relevant 167 information, but some analysis results are included in the appendix of the article. 168 169 170 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 7 Table 2. Textural properties of the materials employed in the solid bed. 171 CSi Coke Refractory Surface area (m2 g-1) 0.085 1.18 0.28 Pore volume (cm3 g-1) 4.18·10-4 3.48·10-3 6.15·10-4 Pore diameter (Å) 34.8 33.9 55.8 172 2.4. Analytical techniques 173 The carbon fiber content of the CFRP was determined using the ASTM D3171 (Procedure B), 174 which consists of the chemical dissolution of the resin in a mixture of sulfuric acid and hydrogen 175 peroxide. The ultimate analysis was determined with the LECO TrueSpec CHN and LECO 176 TrueSpec S elemental analyzers respectively. Additionally, the presence of elemental chlorine 177 was examined by applying Method 5050 of the US Environmental Protection Agency (EPA). This 178 method is based on the combustion of the sample in a calorimetric pump (LECO AC-500) and 179 the subsequent absorption of the combustion gases in a basic solution, which is analyzed by ion 180 chromatography (Dionex ICS 3000, equipped with conductivity and amperometric detectors). 181 The proximate analysis was performed with the LECO TGA-500 thermobalance, which follows 182 the ASTM D3173-85 and D3174-82 standards. 183 184 The textural properties of the solid bed materials was determined using the N2 adsorption185 desorption method in an Autosorb 1C-TCD. The samples were previously degassed at 300 °C for 186 12 h. The surface area was calculated using the Brunauer-Emmett-Teller (BET) method, while 187 the pore distribution was calculated using the Barrett-Joyner-Halenda (BJH) method. The acidity 188 of the refractory material was analyzed by temperature programmed desorption of ammonia 189 (NH3-TPD) while the reducible metals were sought by temperature-programmed reduction 190 (TPR), both of them by using the Micromeritics® AutoChem II equipment. This equipment was 191 also used to analyze the presence of catalytically active metals on the surface of refractory by 192 chemisorption with CO. 193 194 The pyrolysis solids were characterized through ultimate and proximate analysis, following the 195 procedure used with the CFRP sample. The composition of the liquid products was determined 196 using a gas chromatograph (GC, Agilent 6890) coupled to a mass spectrometer (MS, Agilent 197 5973), considering as identified only such chemicals providing an identification quality 198 parameter greater than 85%. The gaseous products were analyzed in a gas chromatograph 199 (Agilent 7890A) with thermal conductivity and flame ionization detectors (TCD and FID, 200 respectively). This chromatograph is calibrated to determine the composition of the so-called 201 "refinery mixture", which includes H2, CO, CO2, CH4, C2, C3, C4, C5 and C6. The higher heating 202 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 8 value (HHV) of the gases was calculated from the HHV of the pure compounds that are part of 203 their composition. 204 205 The reclaimed CF were also subjected to ultimate and proximate analyses according to the 206 procedure mentioned above. The morphological characteristics were evaluated by scanning 207 electron microscopy (SEM), with energy-dispersive X-ray spectroscopy (EDX) for the 208 determination of the surface composition, using the Hitachi S-2400N microscope. Mechanical 209 properties were measured by means of Minimat 2000 tester (Rheometric Scientific) equipped 210 with a load cell of 200 N at a testing speed of 1 mm min-1. CF were mounted and glued on a 211 paper tab that was placed in the grips. Then, the middle portion of the papers was cut, so that 212 only the carbon fiber carried the load (Luo and Netravali, 2001). 10 mm clamping length was 213 used and at least 10 specimens were tested, being the average values reported. 214 215 3. Results and discussion 216 217 3.1. Pyrolysis stage 218 Table 3 shows the pyrolysis yields and the composition of the gases and liquids obtained in the 219 five tests. These pyrolysis yields are the average value of at least two identical experiments 220 where there was not a variation greater than 3% in the yields themselves. These yields are 221 quantified as follows: (1) solids yield corresponds to the remaining sample in the reactor after 222 the pyrolysis stage, (2) the yield of total condensates is determined by the difference in weight 223 of the pipes and condensers before and after the experiment, and (3) the gas yield is calculated 224 by difference. The "collected liquids" is the amount of liquids in the condensers, which is 225 distinguished within the total condensates yield with the intention of giving representative 226 information of the process parameters and not so much of the geometry of the installation, 227 which can have much influence on the amount of condensates generated in the pipes and other 228 elements. These collected liquids have been used to determine the composition of condensates 229 and the proportions of aqueous phase and organic phase. The gas composition shown in the 230 table is dry and free of N2. 231 232 233 234 235 236 237 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 15 398 Figure 3. SEM images of reclaimed carbon fibers. 399 400 Energy-dispersive X-ray spectroscopy (EDX) has been used to determine the surface composition 401 of the CF, as well as that of the particles seen on SEM micrographs. The results are presented in 402 Figures 4 and 5. Firstly, one smooth area has been analyzed, that is, the surface of the fibers 403 (Figure 4). The spectrogram shows that the only element detected on that surface is carbon, 404 which could indicate that it is the surface of the clean fiber, and therefore would indicate a good 405 reclamation of the material. Previous research on oxidative treatments of CFRP waste has also 406 shown small amounts of oxygen on the surface of the reclaimed fibers (Yip et al., 2002). In 407 general, the amount of surface oxygen increases with the aggressiveness of the oxidation 408 process (long time, high temperature), so the absence of surface oxygen could be indicative of 409 controlled and not excessive oxidation. 410 411 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 16 412 Figure 4. Carbon fibers surface composition analysis. 413 414 Figure 5 shows the spectrograms of two particles deposited on the fibers. It can be seen that the 415 main element of their composition is again carbon, followed by oxygen, sodium and sulphur. 416 The presence of oxygen and sulphur may indicate some traces of the polymeric resin, while the 417 appearance of sodium on the surface is a priori quite surprising. However, there is a precedent 418 reporting the presence of sodium in reclaimed fibers from CFRP. In that work, the existence of 419 sodium was attributed to the manufacturing process of the carbon fibers (Jiang et al., 2008). 420 421 422 423 Figure 5. Composition of surface particles. 424 425 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 17 The mechanical characterization of the reclaimed fibers is presented in Table 5. The parameters 426 measured are the usual ones in mechanical characterization of this material, that is, tensile 427 strength, tensile modulus and elongation at break. The values of the same parameters of some 428 commercial CF have been also included in this table, obtained from commercial catalogues and 429 available technical data sheets. It can be seen that the recovered fibers have a considerably 430 lower tensile modulus than the commercial fibers of the TORAYCA and the YS series of GRANOC, 431 all of them common in structural aeronautical applications. However, the tensile strength, 432 although also lower, is in the order of magnitude of the TORAYCA T300 and GRANOC YS-90A433 30S fibers. 434 435 Table 5. Mechanical properties of reclaimed carbon fibers and some commercial carbon fibers. 436 437 The CFRP waste used in this work has been generated in production tasks of the aeronautical 438 industry, but both its specific application and the properties of the original fibers are unknown, 439 so it is not possible to determine the degree of deterioration that the mechanical properties of 440 the fibers may have suffered during the recycling process. However, the conservation of the 441 mechanical properties of CF recovered by pyrolysis is a fact that is proven in the scientific 442 literature related to CFRP recycling, as discussed in the introduction of this article (Naqvi et al., 443 2018; Pimenta and Pinho, 2014, 2011). The low values observed in this study may be due to the 444 specific characteristics of the reactor used or to the uncertainty and dispersion of the results 445 obtained in the analysis of such small diameter fibers, as some researchers have pointed out 446 (Gillet et al., 2015). 447 448 In any case, the elongation at break and tensile modulus of the reclaimed fibers are equivalent 449 to those of the GRANOC XN series fibers, which are classified as low modulus and high elongation 450 fibers. In addition, their tensile strength is significantly higher than the strength of these 451 commercial fibers. Therefore, it is clear that the CF recovered in this work can have value and 452 Tensile strength (MPa) Tensile modulus (GPa) Elongation at break (%) Reclaimed fibers 2678 ± 700 96 ± 20 2.6 ± 0.8 TORAYCA T300 3530 230 1.5 TORAYCA T800S 5880 294 2.0 GRANOC YS-90A-30S 3530 880 0.3 GRANOC XN-10-30-S 1700 110 1.6 GRANOC XN-05-30-S 1100 54 2.0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 18 commercial applications and compete with virgin carbon fibers of fossil origin. These results 453 demonstrate the competitiveness of reclaimed carbon fibers as secondary raw materials. 454 455 4. Conclusion 456 457 The main conclusions that can be drawn from the work presented in this article are the 458 following: 459 460 - Pyrolysis and oxidation of poly(benzoxazine) type CFRP waste allows the recovery of carbon 461 fibers that present mostly clean surfaces and mechanical properties similar to some commercial 462 virgin carbon fibers. 463 - The cooling and condensation of the vapors from the pyrolysis stage generates a liquid and a 464 gaseous fraction. The liquid fraction is composed of an aqueous and an organic phase, both of 465 which are of low value due to their complex composition based on nitrogenous and oxygenated 466 aromatic compounds. The gaseous fraction is composed of CH4 and CO2, it has low calorific value 467 and is generated in small quantities. 468 - If the vapors are treated at high temperature (900 °C) in a CSi-filled reactor, the amount of 469 liquids generated is significantly reduced, increasing the proportion of gases to more than 470 double. These gases improve their composition significantly, enriching themselves with H2 and 471 reducing their CO2 content, which doubles their calorific value. 472 - It has been proven that the material used as solid bed in the tubular reactor has an appreciable 473 influence. The best results have been obtained by using a residual refractory material, thanks to 474 which the profitability of the process increases considerably. This is why: (1) it is a waste material 475 and (2) it generates a H2 proportion higher than 50% by volume in the gases, which enables the 476 separation of H2 as a pure compound, presenting a wide range of applications. 477 - The results obtained in this work show that the material recovering of the polymeric resin in 478 the CFRP waste recycling process is possible, which represents a considerable advance in the 479 field. This is got by a technically simple and scalable method, very usual in the chemical industry. 480 As far as costs are concerned, the only relevant cost is the energy consumption to reach the 481 treatment temperature, since solid bed material is a reusable waste material. Besides, this 482 energy could be obtained from the gases through the separation of H2 by pressure swing 483 adsorption (PSA) and the employment of the by-product gas, which would retain high HHV, to 484 generate that energy. 485 486 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 19 Acknowledgements 487 The authors want to thank the Ministry of Economy and Competitiveness of the Spanish 488 Government for the funding (Ref. CTM2013-48887-C2-1-R). The authors also thank the Basque 489 Government for the Researcher Training Grant awarded to N. Gastelu in 2015 call and the 490 financing granted to the “Sustainable Process Engineering” research group for the 2016-2021 491 period. 492 493 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 20 References 494 Adrados, A., Lopez-Urionabarrenechea, A., Solar, J., Requies, J., De Marco, I., Cambra, J.F., 495 2013. Upgrading of pyrolysis vapours from biomass carbonization. J. Anal. Appl. Pyrolysis 496 103, 293–299. https://doi.org/10.1016/j.jaap.2013.03.002 497 Cousins, D.S., Suzuki, Y., Murray, R.E., Samaniuk, J.R., Stebner, A.P., 2019. Recycling glass fiber 498 thermoplastic composites from wind turbine blades. J. Clean. Prod. 209, 1252–1263. 499 https://doi.org/10.1016/J.JCLEPRO.2018.10.286 500 Cunliffe, A.M., Jones, N., Williams, P.T., 2003. Recycling of fibre-reinforced polymeric waste by 501 pyrolysis: thermo-gravimetric and bench-scale investigations. J. Anal. Appl. Pyrolysis 70, 502 315–338. https://doi.org/10.1016/S0165-2370(02)00161-4 503 Dey, A., Kayal, N., Molla, A.R., Chakrabarti, O., 2014. Investigation of thermal oxidation of 504 Al2O3-coated SiC powder. Thermochim. Acta 583, 25–31. 505 https://doi.org/10.1016/J.TCA.2014.03.011 506 Elkasabi, Y., Liu, Q., Choi, Y.S., Strahan, G., Boateng, A.A., Regalbuto, J.R., 2017. Bio-oil 507 hydrodeoxygenation catalysts produced using strong electrostatic adsorption. Fuel 207, 508 510–521. https://doi.org/10.1016/J.FUEL.2017.06.115 509 Gastelu, N., Lopez-Urionabarrenechea, A., Acha, E., Caballero, B.M., de Marco, I., 2019. 510 Evaluation of HZSM-5 Zeolite as Cracking Catalyst for Upgrading the Vapours Generated 511 in the Pyrolysis of an Epoxy-Carbon Fibre Waste Composite. Top. Catal. 512 https://doi.org/10.1007/s11244-019-01152-7 513 Gastelu, N., Lopez-Urionabarrenechea, A., Solar, J., Acha, E., Caballero, B.M., López, F.A., de 514 Marco, I., 2018. Thermo-Catalytic Treatment of Vapors in the Recycling Process of Carbon 515 Fiber-Poly (Benzoxazine) Composite Waste by Pyrolysis. Catalysts 8, 523. 516 https://doi.org/10.3390/catal8110523 517 Ghosh, N.N., Kiskan, B., Yagci, Y., 2007. Polybenzoxazines—New high performance 518 thermosetting resins: Synthesis and properties. Prog. Polym. Sci. 32, 1344–1391. 519 https://doi.org/10.1016/J.PROGPOLYMSCI.2007.07.002 520 Gillet, A., Mantaux, O., Cazaurang, G., 2015. Characterization of composite materials made 521 from discontinuous carbon fibres within the framework of composite recycling. Compos. 522 Part A Appl. Sci. Manuf. 75, 89–95. https://doi.org/10.1016/J.COMPOSITESA.2015.05.002 523 Giorgini, L., Benelli, T., Mazzocchetti, L., Leonardi, C., Zattini, G., Minak, G., Dolcini, E., 524 Cavazzoni, M., Montanari, I., Tosi, C., 2015. Recovery of carbon fibers from cured and 525 uncured carbon fiber reinforced composites wastes and their use as feedstock for a new 526 composite production. Polym. Compos. 36, 1084–1095. 527 https://doi.org/10.1002/pc.23440 528 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 21 Guo, F., Li, X., Liu, Y., Peng, K., Guo, C., Rao, Z., 2018. Catalytic cracking of biomass pyrolysis tar 529 over char-supported catalysts. Energy Convers. Manag. 167, 81–90. 530 https://doi.org/10.1016/J.ENCONMAN.2018.04.094 531 Hermansson, F., Janssen, M., Svanström, M., 2019. Prospective study of lignin-based and 532 recycled carbon fibers in composites through meta-analysis of life cycle assessments. J. 533 Clean. Prod. 223, 946–956. https://doi.org/10.1016/J.JCLEPRO.2019.03.022 534 Holmes, M., 2018. Recycled carbon fiber composites become a reality. Reinf. Plast. 62, 148– 535 153. https://doi.org/10.1016/J.REPL.2017.11.012 536 Holmes, M., 2017a. Carbon composites continue to find new markets. Reinf. Plast. 61, 36–40. 537 https://doi.org/10.1016/J.REPL.2016.12.060 538 Holmes, M., 2017b. Lowering the cost of carbon fiber. Reinf. Plast. 61, 279–283. 539 https://doi.org/10.1016/J.REPL.2017.02.001 540 Hu, M., Laghari, M., Cui, B., Xiao, B., Zhang, B., Guo, D., 2018. Catalytic cracking of biomass tar 541 over char supported nickel catalyst. Energy 145, 228–237. 542 https://doi.org/10.1016/J.ENERGY.2017.12.096 543 Irisawa, T., Inagaki, R., Iida, J., Iwamura, R., Ujihara, K., Kobayashi, S., Tanabe, Y., 2018. The 544 influence of oxygen containing functional groups on carbon fibers for mechanical 545 properties and recyclability of CFRTPs made with in-situ polymerizable polyamide 6. 546 Compos. Part A Appl. Sci. Manuf. 112, 91–99. 547 https://doi.org/10.1016/J.COMPOSITESA.2018.05.035 548 Jiang, G., Pickering, S.J., Walker, G.S., Wong, K.H., Rudd, C.D., 2008. Surface characterisation of 549 carbon fibre recycled using fluidised bed. Appl. Surf. Sci. 254, 2588–2593. 550 https://doi.org/10.1016/J.APSUSC.2007.09.105 551 Khalil, Y.F., 2018. Comparative environmental and human health evaluations of thermolysis 552 and solvolysis recycling technologies of carbon fiber reinforced polymer waste. Waste 553 Manag. 76, 767–778. https://doi.org/10.1016/J.WASMAN.2018.03.026 554 Kim, Y.N., Kim, Y.O., Kim, S.Y., Park, M., Yang, B., Kim, J., Jung, Y.C., 2019. Application of 555 supercritical water for green recycling of epoxy-based carbon fiber reinforced plastic. 556 Compos. Sci. Technol. 173, 66–72. https://doi.org/10.1016/j.compscitech.2019.01.026 557 La Rosa, A.D., Banatao, D.R., Pastine, S.J., Latteri, A., Cicala, G., 2016. Recycling treatment of 558 carbon fibre/epoxy composites: Materials recovery and characterization and 559 environmental impacts through life cycle assessment. Compos. Part B Eng. 104, 17–25. 560 https://doi.org/10.1016/j.compositesb.2016.08.015 561 Lee, J., Kim, K.-H., Kwon, E.E., 2017. Biochar as a Catalyst. Renew. Sustain. Energy Rev. 77, 70– 562 79. https://doi.org/10.1016/J.RSER.2017.04.002 563 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 22 Lefeuvre, A., Garnier, S., Jacquemin, L., Pillain, B., Sonnemann, G., 2017. Anticipating in-use 564 stocks of carbon fiber reinforced polymers and related waste flows generated by the 565 commercial aeronautical sector until 2050. Resour. Conserv. Recycl. 125, 264–272. 566 https://doi.org/10.1016/J.RESCONREC.2017.06.023 567 Liu, P., Meng, F., Barlow, C.Y., 2019. Wind turbine blade end-of-life options: An eco-audit 568 comparison. J. Clean. Prod. 212, 1268–1281. 569 https://doi.org/10.1016/J.JCLEPRO.2018.12.043 570 Liu, Y., Farnsworth, M., Tiwari, A., 2017. A review of optimisation techniques used in the 571 composite recycling area: State-of-the-art and steps towards a research agenda. J. Clean. 572 Prod. 140, 1775–1781. https://doi.org/10.1016/J.JCLEPRO.2016.08.038 573 Longana, M.L., Ong, N., Yu, H., Potter, K.D., 2016. Multiple closed loop recycling of carbon fibre 574 composites with the HiPerDiF (High Performance Discontinuous Fibre) method. Compos. 575 Struct. 153, 271–277. https://doi.org/10.1016/J.COMPSTRUCT.2016.06.018 576 López, A., de Marco, I., Caballero, B.M., Laresgoiti, M.F., Adrados, A., Aranzabal, A., 2011. 577 Catalytic pyrolysis of plastic wastes with two different types of catalysts: ZSM-5 zeolite 578 and Red Mud. Appl. Catal. B Environ. 104, 211–219. 579 https://doi.org/10.1016/j.apcatb.2011.03.030 580 López, F.A., Rodríguez, O., Alguacil, F.J., García-Díaz, I., Centeno, T.A., García-Fierro, J.L., 581 González, C., 2013. Recovery of carbon fibres by the thermolysis and gasification of waste 582 prepreg. J. Anal. Appl. Pyrolysis 104, 675–683. 583 https://doi.org/10.1016/J.JAAP.2013.04.012 584 Luo, S., Netravali, A.N., 2001. Characterization of henequen fibers and the henequen 585 fiber/poly(hydroxybutyrate-co-hydroxyvalerate) interface. J. Adhes. Sci. Technol. 15, 586 423–437. https://doi.org/10.1163/156856101300157533 587 Mazzocchetti, L., Benelli, T., D’Angelo, E., Leonardi, C., Zattini, G., Giorgini, L., 2018. Validation 588 of carbon fibers recycling by pyro-gasification: The influence of oxidation conditions to 589 obtain clean fibers and promote fiber/matrix adhesion in epoxy composites. Compos. 590 Part A Appl. Sci. Manuf. 112, 504–514. 591 https://doi.org/10.1016/J.COMPOSITESA.2018.07.007 592 Meng, F., McKechnie, J., Pickering, S.J., 2018. An assessment of financial viability of recycled 593 carbon fibre in automotive applications. Compos. Part A Appl. Sci. Manuf. 109, 207–220. 594 https://doi.org/10.1016/J.COMPOSITESA.2018.03.011 595 Meyer, L.O., Schulte, K., Grove-Nielsen, E., 2009. CFRP-Recycling Following a Pyrolysis Route: 596 Process Optimization and Potentials. J. Compos. Mater. 43, 1121–1132. 597 https://doi.org/10.1177/0021998308097737 598 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 23 Nahil, M.A., Williams, P.T., 2011. Recycling of carbon fibre reinforced polymeric waste for the 599 production of activated carbon fibres. J. Anal. Appl. Pyrolysis 91, 67–75. 600 https://doi.org/10.1016/J.JAAP.2011.01.005 601 Naqvi, S.R., Prabhakara, H.M., Bramer, E.A., Dierkes, W., Akkerman, R., Brem, G., 2018. A 602 critical review on recycling of end-of-life carbon fibre/glass fibre reinforced composites 603 waste using pyrolysis towards a circular economy. Resour. Conserv. Recycl. 136, 118–129. 604 https://doi.org/10.1016/J.RESCONREC.2018.04.013 605 Newcomb, B.A., 2016. Processing, structure, and properties of carbon fibers. Compos. Part A 606 Appl. Sci. Manuf. 91, 262–282. https://doi.org/10.1016/J.COMPOSITESA.2016.10.018 607 Onwudili, J.A., Miskolczi, N., Nagy, T., Lipóczi, G., 2016. Recovery of glass fibre and carbon 608 fibres from reinforced thermosets by batch pyrolysis and investigation of fibre re-using as 609 reinforcement in LDPE matrix. Compos. Part B Eng. 91, 154–161. 610 https://doi.org/10.1016/J.COMPOSITESB.2016.01.055 611 Ortega, C., Rezaei, M., Hessel, V., Kolb, G., 2018. Methanol to dimethyl ether conversion over a 612 ZSM-5 catalyst: Intrinsic kinetic study on an external recycle reactor. Chem. Eng. J. 347, 613 741–753. https://doi.org/10.1016/J.CEJ.2018.04.160 614 Pimenta, S., Pinho, S.T., 2014. Recycling of Carbon Fibers. Handb. Recycl. 269–283. 615 https://doi.org/10.1016/B978-0-12-396459-5.00019-2 616 Pimenta, S., Pinho, S.T., 2011. Recycling carbon fibre reinforced polymers for structural 617 applications: Technology review and market outlook. Waste Manag. 31, 378–392. 618 https://doi.org/10.1016/J.WASMAN.2010.09.019 619 Ravenni, G., Sárossy, Z., Ahrenfeldt, J., Henriksen, U.B., 2018. Activity of chars and activated 620 carbons for removal and decomposition of tar model compounds – A review. Renew. 621 Sustain. Energy Rev. 94, 1044–1056. https://doi.org/10.1016/J.RSER.2018.07.001 622 Rybicka, J., Tiwari, A., Leeke, G.A., 2016. Technology readiness level assessment of composites 623 recycling technologies. J. Clean. Prod. 112, 1001–1012. 624 https://doi.org/10.1016/J.JCLEPRO.2015.08.104 625 Solar, J., Caballero, B., De Marco, I., López-Urionabarrenechea, A., Gastelu, N., 2018. 626 Optimization of Charcoal Production Process from Woody Biomass Waste: Effect of Ni627 Containing Catalysts on Pyrolysis Vapors. Catalysts 8, 191. 628 https://doi.org/10.3390/catal8050191 629 Song, H., Zhao, Q., Zhou, X., Cao, Z., Luo, M., 2018. Selection of highly active and stable Co 630 supported SiC catalyst for Fischer-Tropsch synthesis: Effect of the preparation method. 631 Fuel 229, 144–150. https://doi.org/10.1016/J.FUEL.2018.05.025 632 Sun, K., Huang, Q., Chi, Y., Yan, J., 2018. Effect of ZnCl2-activated biochar on catalytic pyrolysis 633 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 24 of mixed waste plastics for producing aromatic-enriched oil. Waste Manag. 81, 128–137. 634 https://doi.org/10.1016/J.WASMAN.2018.09.054 635 Tillmann, L., Schulwitz, J., van Veen, A., Muhler, M., 2018. Dry Reforming of Methane at High 636 Pressure in a Fixed-Bed Reactor with Axial Temperature Profile Determination. Catal. 637 Letters 148, 2256–2262. https://doi.org/10.1007/s10562-018-2453-x 638 Tranchard, P., Duquesne, S., Samyn, F., Estèbe, B., Bourbigot, S., 2017. Kinetic analysis of the 639 thermal decomposition of a carbon fibre-reinforced epoxy resin laminate. J. Anal. Appl. 640 Pyrolysis 126, 14–21. https://doi.org/10.1016/J.JAAP.2017.07.002 641 Vo Dong, A., Azzaro-Pantel, C., Boix, M., 2019. A multi-period optimisation approach for 642 deployment and optimal design of an aerospace CFRP waste management supply chain. 643 Waste Manag. 95, 201–216. https://doi.org/10.1016/J.WASMAN.2019.05.002 644 Vo Dong, P.A., Azzaro-Pantel, C., Cadene, A.-L., 2018. Economic and environmental assessment 645 of recovery and disposal pathways for CFRP waste management. Resour. Conserv. Recycl. 646 133, 63–75. https://doi.org/10.1016/J.RESCONREC.2018.01.024 647 Wang, F.C.-Y., Dettloff, M.., Null, M.., White, J.., 2000. Qualitative and quantitative analysis of a 648 thermoset polymer, poly(benzoxazine), by pyrolysis–gas chromatography. J. Chromatogr. 649 A 886, 217–224. https://doi.org/10.1016/S0021-9673(00)00508-2 650 Wang, N., Chen, D., Arena, U., He, P., 2017. Hot char-catalytic reforming of volatiles from MSW 651 pyrolysis. Appl. Energy 191, 111–124. https://doi.org/10.1016/J.APENERGY.2017.01.051 652 William D. Callister Jr., D.G.R., 2014. Materials Science and Engineering: An introduction, 9th 653 ed. John Wiley & Sons, Ltd, New Jersey (USA). 654 Wong, K.H., Pickering, S.J., Rudd, C.D., 2010. Recycled carbon fibre reinforced polymer 655 composite for electromagnetic interference shielding. Compos. Part A Appl. Sci. Manuf. 656 41, 693–702. https://doi.org/10.1016/J.COMPOSITESA.2010.01.012 657 Xu, X., Li, Z., Tu, R., Sun, Y., Jiang, E., 2018. Hydrogen from Rice Husk Pyrolysis Volatiles via 658 Non-Noble Ni–Fe Catalysts Supported on Five Differently Treated Rice Husk Pyrolysis 659 Carbon Supports. ACS Sustain. Chem. Eng. 6, 8325–8339. 660 https://doi.org/10.1021/acssuschemeng.8b00343 661 Yang, J., Liu, J., Liu, W., Wang, J., Tang, T., 2015. Recycling of carbon fibre reinforced epoxy 662 resin composites under various oxygen concentrations in nitrogen–oxygen atmosphere. J. 663 Anal. Appl. Pyrolysis 112, 253–261. https://doi.org/10.1016/J.JAAP.2015.01.017 664 Yang, Y., Boom, R., Irion, B., van Heerden, D.-J., Kuiper, P., de Wit, H., 2012. Recycling of 665 composite materials. Chem. Eng. Process. Process Intensif. 51, 53–68. 666 https://doi.org/10.1016/J.CEP.2011.09.007 667 Yip, H.L.H., Pickering, S.J., Rudd, C.D., 2002. Characterisation of carbon fibres recycled from 668 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65