Stochastic consequential Life Cycle Assessment of technology substitution in the case of a novel PET chemical recycling technology

The current traditional mechanical recycling of Polyethylene Terephthalate (PET) grinds the waste into granulate,yet the resulting secondary material quality is strongly dependent on the efficiency of the selection processes,leading to the requirement of an integration of fossil PET to assure the bottle-grade quality is reached.Instead, the novel chemical recycling gr3n technology depolymerizes the waste PET back into the constituentmonomers with a resulting quality that is comparable to the virgin product, due to a more efficient separation ofimpurities. In order to estimate the environmental impacts related to the introduction of this technology in therelated market, Consequential Life Cycle Assessment (CLCA) is particularly indicated. Among the consequentialapproaches, we adopt the Stochastic Technology Choice Model, as it is able to model the technological mixestypical of markets based on costs and production capacities, while its stochasticity suits the need to manage theuncertainty of future market conditions. Indeed, the assessment of the expected technological mixes contributingto the same function and the quality of the recycled material are key to evaluate the variation in marginal LCAimpacts due to the introduction of the gr3n technology. We assess the marginal LCA impacts of the Europeanbottle-grade PET market in two scenarios: one in which the gr3n technology is not available and one in whichthis technology is present. To correctly evaluate the difference between these two scenarios, we perform a pairedsimulation. Here we show that the populations related to this difference show more than 50% negative results in12 out of 16 impact indicators and more than 75% of negative results in 9 out of 16 impact indicators. Inparticular, a median 0.13 kg CO2-eq per kg bottle-grade PET could be saved by the introduction of gr3n, equivalent toa 5% reduction. We show that the 5-95 percentiles range of the difference between the two scenarios is only17.7% of the average range defined by the two separate scenarios distributions, confirming previous findingsfrom the literature. The robustness of the results is tested through three sensitivity analyses. Therefore, policymakers should focus on limiting the increase in marginal demand of PET and on creating fair conditions for thischemical recycling technology to be deployed to complement mechanical recycling in reducing virgin PETproduction, thus decreasing potential environmental impacts and fostering a more circular economy. The positiveperformance of the novel technology is strongly related to the increased substitution of waste treatmentprocesses, such as incineration and landfill, and to the increased quality of the recycled product: this environmentalprofile could further improve as the novel technology will scale up industrially.