ABSTRACT
Differential scanning calorimetry (DSC), a routine thermoanalytical method in material science, is gaining utility in plastic pollution research to improve polymer identification. We optimized a DSC method, experimentally testing pan types, temperature ramps, number of melts, and minimum sample masses. Using the optimized method, we created an in-house thermogram library from 201 polymer reference standards. We determined peak melting temperature cutoffs for differentiating variants of PE and nylon. PE cutoffs remained stable after experimentally weathering standards outdoors or for severely weathered HDPE debris found on Hawaii's beaches. Marine debris samples, across a range of weathering severity and previously identified as either low-density or high-density polyethylene (LDPE or HDPE) based on the 1377 cm-1 peak indicating methyl groups by attenuated total reflectance Fourier-transform infrared spectroscopy (ATR-FTIR), were analyzed by DSC to confirm or challenge the ATR-FTIR PE differentiation. ATR-FTIR was correct for >80% of the HDPE samples, but <40% of those initially identified as LDPE by ATR-FTIR. Accuracy did not relate to weathering extent. Most samples mis-identified as LDPE were HDPE that had formed methyl groups likely from chain scission during photooxidation. ATR-FTIR alone is unreliable for differentiating weathered PE, DSC is required. We provide a multiple-method workflow for complete and accurate polymer identification, even for microplastics ≥0.03 mg. Applying these methods can better identify the polymer composition of marine debris, essential for sourcing and recycling efforts.
