An Anticipatory Approach to Improve the Recycling of CdTe Photovoltaic Systems

The waste from end-of-life photovoltaic (PV) systems is projected to reach 78 million tons by 2070 and this necessitates environmentally improved recycling processes. The incumbent industrial-scale CdTe recycling process... [ view full abstract ]

The waste from end-of-life photovoltaic (PV) systems is projected to reach 78 million tons by 2070 and this necessitates environmentally improved recycling processes. The incumbent industrial-scale CdTe recycling process requires two steps- eliminating the ethylene vinyl acetate (EVA) polymer encapsulating the glass, and recovering the CdTe layer through mechanical shredding and recovering the cadmium and tellurium through chemical leaching and precipitation. The elimination of EVA is an environmental hotspot and addressing this requires a comparative environmental impact assessment of novel recycling alternatives and the incumbent process. Lifecycle Assessment (LCA), the preferred framework for environmental decision making, requires material and energy inventory data from industry processes that mature over a time-period and, therefore, is retrospective. This methodologically limits the application of LCA to evaluate novel pilot-scale processes with limited or no inventory data and prospectively guide them towards environmentally favorable pathways.

We address this limitation by applying the novel anticipatory-LCA (a-LCA) framework to compare and environmentally rank 6 novel lab-scale and the incumbent industrial recycling processes based on 18 environmental impact categories. The inventory data for the a-LCA of three novel processes was collected from laboratory experiments and three other processes from secondary literature. Also, we determine if decentral or decentral recycling is favorable for managing PV waste in the future, identify the most significant parameters that impact the environmental ranking, and determine the key uncertainties to be resolved for increasing the confidence in the choice of the preferred method.

The results show that in the first step of EVA elimination, thermal delamination at 500C is environmentally preferable to the incumbent process and the novel processes of ultrasonication and dissolution in an organic solvent. In the second step, the recovery of cadmium and tellurium through acidic leaching and precipitation is environmentally preferable to the novel alternatives of ion-exchange and solvent extraction. Decentralized recycling in mobile plants at the deployment site is environmentally preferred when road is the dominant mode of transporting end-of-life modules from deployment site to the centralized recycling location. Similarly, centralized recycling is environmentally favorable when shipping is the dominant mode of transportation. A global sensitivity analysis (GSA) shows that weights allocated to the 18 environmental impact categories is the most significant parameter influencing the environmental ranking of the recycling alternatives. This suggests that sensitivity of the stakeholder to an environmental impact significantly influences the choice of the recycling process. The electricity footprint is the second most important parameter and recyclers can lower the electricity required or shift to carbon-efficient sources of electricity to improve the performance of the novel thermal delamination process in the future. Furthermore, research is required to identify the contributors to the uncertainties in the climate change, fossil fuel depletion and the ozone depletion environmental impact categories to improve the confidence in the environmental ranking and better inform the stakeholder on the environmental trade-offs in choosing between the thermal delamination and the incumbent commercial-scale recycling process.