(578a) Modeling and Evaluation of Pyrolysis and Releases to Track the Environmental Fate of Chemicals of Concern

Over 2 billion tons of municipal solid waste (MSW) are generated annually with an estimated increase to 3.4 billion annually by 2050 [1]. Considering that plastics contribute approximately 7% to 12% of the total MSW by weight, their end-of-life (EoL) stage management remains a significant challenge. It is estimated that only 9% to 10% of the annual plastic production is recycled, 10% is incinerated, while around 80% is discarded without being reused or repurposed [1]. Therefore, plastic pollution is a concerning problem which has several impacts on the environment and human health. Pyrolysis is a well-studied technique for converting EoL plastic into valuable products, which involves thermal decomposition in an oxygen-free environment. The main goal is producing liquid fuels, but it can also be used for obtaining specific monomers or chemicals [2]. Besides, plastics can contain a variety of chemical substances, including residual monomers from polymer production, additives designed to enhance specific properties, processing aids, and non-intentionally added substances. Studies identified more than 16,000 substances, of which more than 4,200 are chemicals of concern, meaning that they are persistent, bio-accumulative, mobile, and/or toxic [3]. Tracking the environmental fate of these chemicals presents a significant challenge; however, an even greater difficulty lies in identifying and monitoring the compounds produced and released during the pyrolysis of plastics. Thus, understanding the transformation and release of these chemicals during pyrolysis processes is essential for assessing potential risks and developing safer EoL management.

Mechanistic modeling of polymer degradation has effectively been applied to understand the influence of time, temperature, pressure, and feedstock variability on product distribution based on molecular structure. Stochastic chemical kinetics, specifically, Kinetic Monte Carlo (KMC) modeling has been used to provide detailed and explicit information about the distribution and composition of the polymer chains [4]. KMC accounts for the discrete nature of molecules and the randomness in their behavior, which is useful in systems at a molecular scale. In this process, simulation typically involves evaluating the reaction rates for all possible reactions at a given state of the system; these rates are then used to determine both the timing of the next reaction and which reaction will take place; once a reaction occurs, the system state is updated accordingly, and the process is repeated iteratively to model the systems’ behavior over time [5]. In addition, generative Artificial Intelligence (AI) can enhance kinetic modeling as it helps identify potential reaction pathways and molecular structures that can be produced, thus ensuring that the models are practically applicable, and the molecules produced are tractable [6]. Thus, in this work we integrate the knowledge of plastics, recycling processes, and apply KMC and generative AI to identify and monitor the releases of chemicals of concern during the pyrolysis process based on an initial set of possible reactions and the molecular structure of the compounds. We anticipate that this shall aid in understanding the transformation and release of chemicals of concern and tracking their environmental fate during the pyrolysis of EoL plastic. This holistic material flow analysis supports advancements in estimating environmental releases and exposure pathways of chemicals during the pyrolysis of EoL materials. Our technical findings will also aid the development of potentially significant new use rules (SNURs) for new uses of EoL materials as energy and feedstock sources and evaluate chemical risks.

Disclaimer: The views expressed in this abstract are those of the authors and do not necessarily represent the views or policies of the EPA.

References

[1] K. O. Babaremu et al., “Sustainable plastic waste management in a circular economy,” Heliyon, vol. 8, no. 7, p. e09984, Jul. 2022, doi: 10.1016/j.heliyon.2022.e09984.

[2] M. S. Qureshi et al., “Pyrolysis of plastic waste: Opportunities and challenges,” J. Anal. Appl. Pyrolysis, vol. 152, p. 104804, Nov. 2020, doi: 10.1016/j.jaap.2020.104804.

[3] H. Wiesinger et al., “LitChemPlast : An Open Database of Chemicals Measured in Plastics,” Environ. Sci. Technol. Lett., vol. 11, no. 11, pp. 1147–1160, Nov. 2024, doi: 10.1021/acs.estlett.4c00355.

[4] L. J. Broadbelt et al., “Developing Strategies for Polymer Redesign and Recycling Using Reaction Pathway Analysis,” presented at the Foundations of Computer Aided Process Operations/Chemical Process Control (FOCAPO/CPC), San Antonio, Texas, Jan. 2023. Accessed: Mar. 21, 2025. [Online]. Available: https://skoge.folk.ntnu.no/prost/proceedings/focapo-cpc-2023/Plenary%20…

[5] D. T. Gillespie, “Stochastic Simulation of Chemical Kinetics,” Annu. Rev. Phys. Chem., vol. 58, no. 1, pp. 35–55, May 2007, doi: 10.1146/annurev.physchem.58.032806.104637.

[6] W. Gao, S. Luo, and C. W. Coley, “Generative Artificial Intelligence for Navigating Synthesizable Chemical Space,” Oct. 04, 2024, arXiv: arXiv:2410.03494. doi: 10.48550/arXiv.2410.03494.