Polymer characterization of submerged plastic litter from Lake Tahoe, United States

Monitoring plastic litter in the environment is critical to understanding the amount, sources, transport, fate, and environmental impact of this pollutant. However, few studies have monitored plastic litter on lakebeds which are potentially important environments for determining the fate and transport of plastic litter in freshwater basins. In this study, self-contained underwater breathing apparatus (SCUBA) was used for litter collection at the lakebed along transects at five locations in Lake Tahoe, United States. Litter was brought to the surface and characterized by litter type. Plastic litter was subsampled for polymer determination using Attenuated Total Reflectance-Fourier Transform InfraRed (ATR-FTIR) spectroscopy. The average plastic litter recovered from the lakebed for the five dive sites was 83 ±49 items per kilometer. The top plastic litter categories were other plastic litter (i.e. plastic litter that did not fall in another category), followed by food containers, plastic bags, toys, and bottles <2 L. These results are in line with prior studies on submerged litter, supporting that intervention approaches or ongoing education is needed. The six polymers most frequently detected in the subsamples were polyvinyl chloride, polystyrene/expanded polystyrene, polyethylene terephthalate/polyester, polyethylene, polypropylene, and polyamide. These observations generally reflect the global plastic production trend and global compilations of microplastic studies from lake surface water and sediments. We found that some litter subcategories were primarily comprised of a single polymer type, therefore, in studies where the polymer type cannot be measured but litter is categorized, these results could provide an estimate of the total polymer composition for select subcategories.

Methods
The polymer of each subsample (n=516) was determined through Attenuated Total Reflectance-Fourier Transform InfraRed (ATR-FTIR) spectroscopy using a Thermo Nicolet 6700 FTIR (ThermoFisher, United States) at the University of Nevada, Reno. Spectra were collected using 32 scans with a spectral resolution of 4 cm-1 within the range of 3950 and 650 cm-1. The Norton-Beer strong apodization function was applied by the instrument software (OMNIC software, ThermoFisher, United States) and the background was subtracted from each spectrum automatically by the software. The subtracted background was measured every six spectra (approximately every 10 minutes). The diamond ATR crystal was cleaned with isopropanol/methanol before every background collection. Each sample spectrum was compared to known spectra in the Hummel Polymer Library Database and a database created from the spectra of plastics in the Center for Marine Debris Research Polymer Kit 1.0 (Hawaii Pacific University, United States) collected using the same settings as described above.

In the OMNIC software, we used the correlation approach for comparing the spectra to the libraries. The match value ranges from 0 to 100, with a match of 100 indicating a perfect match. The library match best fitted to the unknown sample spectrum was considered to be the identity of that sample. The polymer match was categorized as “unknown polymer” when the best match was less than a 70% similarity. All matches were additionally verified by the instrument user. For each subsample analyzed, we noted the polymer type and match.