Chemical recycling of End-of-Life wind turbine blades
Description
This report proposes an effective, environmentally-benign, small molecule-assisted, dynamic reaction approach for recycling waste wind turbine blades. At temperatures below 200 °C, thermoset ester-containing resins in waste wind turbine blades and carbon fiber composites were efficiently dissolved via a transesterification reaction. The obtained glass and carbon fiber materials were easily separated from the product solution. Experimental findings demonstrated that this method could recycle a wide range of genuine carbon and glass fiber resin composites, including wind turbine blades constructed from epoxy-anhydride and polyester resin substrates,
Files
Steps to reproduce
Carbon fiber composite sheets (Fig. 1d) and wind turbine blades (supplied as sheets with dimensions of roughly 25 × 30 cm; Fig. 1a) were used as waste samples in the solvolysis process. The WTB samples were trimmed into thin strips (~0.5 cm) to capture the profile of the blade (Fig. 1b). Smaller samples (0.5 × 1 cm) were cut from the strips for solvolysis (Fig. 1c). The CFC sheet was also sliced into thin strips (1.5 × 6.0 cm). The following chemicals were used in the solvolysis process and for resin synthesis from the solvolysis product stream: ethylene glycol (EG; molecular weight [MW] = 62.07 g/mol), 1-methyl-2-pyrrolidinone (NMP; MW = 99.13 g/mol), TBD (MW = 139.20 g/mol), epichlorohydrin (EPI; MW = 92.52 g/mol), isopropanol (i-Pr), sodium hydroxide (NaOH), acetic acid, propylene carbonate (PC; MW = 102.09 g/mol), and glycerol triacetate (TAG; MW = 218.21 g/mol). All reagents were purchased from Sigma-Aldrich (St. Louis, MO, USA) and used as received. Recycling of glass, carbon fiber, and epoxy resin oligomers Chemical recycling was performed at 100-190 °C for 60-180 min under 30-60 bar of an inert N2 atmosphere in a 500 mL Parr 4650 batch reactor (see supplementary material Fig. S1) with the transesterification catalyst, TBD. Ethylene glycol and 1-methyl-2-pyrrolidinone were mixed in a 1:1 ratio to dissolve the catalyst. All tests were carried out in accordance with the experimental design matrix of a central composite design (CCD), and the data were analyzed via analysis of variance (ANOVA) to determine which factors have a statistically significant effect on the resin removal efficiency. Under each set of solvolysis conditions, 1 g of sample and 20 mL of solution were used. To avoid overheating, the Parr 4650 batch reactor was calibrated before each test. To evaluate the scalability of the developed process at atmospheric pressure, we used a three-necked flask with a reflux condenser and a nitrogen supply. The three-necked flask was placed in a temperature-controlled mantle, and 10-50 g of the desired sample and 100-200 mL of ethylene glycol and 1-methyl-2-pyrrolidinone (total) were added. After each experiment, the sample was filtered to extract fiber residues. After filtration, the liquid sample was analyzed by nuclear magnetic resonance spectroscopy (NMR), and the solid residue on the filter was washed, dried, and weighed. After weighing the fiber sample content, it was burned to analyze the contents of the ash. The solvolysis process resin degradation yield (RDY) was determined using Equation: RDY=(w1-w2)/w0 where w1 is the sample weight before solvolysis, w2 is the sample weight after solvolysis, and w0 is the epoxy resin weight before solvolysis.