A generic scenario analysis of end-of-life plastic management: Chemical additives

Abstract

Plastic growing demand and the increment in global plastics production have raised the number of spent plastics, out of which over 90% are either landfilled or incinerated. Both methods for handling spent plastics are susceptible to releasing toxic substances, damaging air, water, soil, organisms, and public health. Improvements to the existing infrastructure for plastics management are needed to limit chemical additive release and exposure resulting from the end-of-life (EoL) stage. This article analyzes the current plastic waste management infrastructure and identifies chemical additive releases through a material flow analysis. Additionally, we performed a facility-level generic scenario analysis of the current U.S. EoL stage of plastic additives to track and estimate their potential migration, releases, and occupational exposure. Potential scenarios were analyzed through sensitivity analysis to examine the merit of increasing recycling rates, using chemical recycling, and implementing additive extraction post-recycling. Our analyses identified that the current state of plastic EoL management possesses high mass flow intensity toward incineration and landfilling. Although maximizing the plastic recycling rate is a reasonably straightforward goal for enhancing material circularity, the conventional mechanical recycling method requires improvement because major chemical additive release and contamination routes act as obstacles to achieving high-quality plastics for future reuse and should be mitigated through chemical recycling and additive extraction. The potential hazards and risks identified in this research create an opportunity to design a safer closed-loop plastic recycling infrastructure to handle additives strategically and support sustainable materials management efforts to transform the US plastic economy from linear to circular.

GRAPHICAL ABSTRACT

1. Introduction

Plastics have proven their use as an essential material in applications ranging from packaging, storage, vehicles, and insulation because of their low cost, versatility, durability, and low weight (Nielsen et al., 2020). However, the current plastic end-of-life (EoL) management pathways are not sustainable and are prone to releasing toxic chemical additives into the surrounding environment (Hahladakis et al., 2018; Hummel, 2002). Without a dramatic shift in the current state of plastics production, usage, and EoL, by the year 2050, the ocean is expected to contain more plastics than fish, while the plastics industry alone will consume 20% of the total oil produced and 15% of the annual carbon budget (World Economic Forum, 2016). Moreover, the presence of plastics in the environment and consumer products may, over time, generate microplastics and nanoplastics that may end up in the digestive tracts of animals and humans (Sridharan et al., 2022). Therefore, the increasing reliance on plastics, in conjunction with the chances of additive migration, requires modification of the existing EoL management pathways to reduce plastic and toxic chemical releases and achieve a circular economy.

Chemical additives release problems are originated from the plastic production stage. Plastics generally require the compounding of polymer resin with various chemicals to achieve the desired properties for a specific application or use (Silviya et al., 2009). The compounding of chemical additives into polymer resins is a physical process in which the additive molecules are not chemically transformed or bound to the polymer matrix. The chemical additives for plastic products include, but are not limited to, antioxidants, antistatic agents, blowing agents, colorants, coupling agents, curing agents, fillers, flame retardants, heat/ultraviolet stabilizers, impact modifiers, lubricants, plasticizers, preservatives, reinforcements, and slip agents. The intended application of plastics determines the combination and amount of chemical additives in the formulation. In most cases, chemical additives are solid powders, flakes, granulates, spheres, and emulsions. These materials can be incorporated into the polymer matrix at various production steps, including polymer production, pelletization, and surface modification (Marturano et al., 2017). The plastic resins can then be converted to finished products for consumer usage (US Environmental Protection Agency (US EPA), 2014a).

As shown in Fig. 1, the plastic wastes found in municipal solid waste (MSW) are capable of releasing chemicals into the surrounding environment based on the following mechanisms: (1) diffusion through the polymer matrix to the surface, (2) desorption from the polymer surface, (3) sorption at the plastic-food interface, and (4) absorption into the surrounding medium (i.e., bulk food phase) (Hahladakis et al., 2018). The chemical diffusion through a polymer is governed by parameters such as pore size, temperature, and external medium (extractant). In some instances, residual monomers and solvents from the manufacturing stage may be present in the polymer matrix. These substances are likely to migrate and evaporate, creating a strong odor. Additives generally have higher molecular weights than solvents and monomers, ranging between 200 and 2000 g/mol (Hansen, 2013). High molecular weight substances are generally large molecules that are not expected to diffuse quickly through the polymer matrix. However, utilizing large additive molecules cannot entirely prevent diffusion because of factors relating to operating temperature and various stimulants serving as the driving force for mass transfer (Hahladakis et al., 2018). The rate of this diffusion is governed by the molecular weight of the substance and the pore size of the plastics. High molecular weight additives can diffuse more slowly than smaller molecules because it has to travel through the pores. Once the additives diffuse to the surface of the polymer, the solubility and compatibility between the two phases may further determine the dispersion of chemical additives throughout the solution. The surrounding medium can influence the migration of chemical additives because of the additive concentration difference between the phases. Chemical additives can slowly diffuse toward the surface of the polymer in order to reach an equilibrium with the surrounding medium. Generally, higher chemical additive migration is observed when plastics are in contact with high temperatures for a long duration and with non-polar substances such as fat and oil (Alin and Hakkarainen, 2011; Ehret-Henry et al., 1994; Fankhauser-Noti and Grob, 2006; Galotto and Guarda, 2004; Gao et al., 2011; Hahladakis et al., 2018; Petersen et al., 1995; Tawfik and Huyghebaert, 1998; Till et al., 1982; Xu et al., 2010).

Fig. 1.

Additive release mechanism of plastics in a medium (food, water supply, landfill). The red object represents chemical additives residing inside the pore of the polymer matrix. The mechanism of releases include (1) diffusion, (2) desorption, (3) sorption, and (4) absorption.

EoL plastic management pathways, such as recycling, incineration, and landfilling, can aid the release of compounded additives into the environment because normal operation only considers the treatment of the plastics rather than the separation between chemical additives and plastics. Mechanical recycling is the most common recycling practice because of its low operational costs and high reliability (Christensen et al., 2019; Kulkarni, 2018; Oladimeji Azeez, 2020; Sherwood, 2020; Wu et al., 2014). The recycling process begins with the separation and sorting of the collected materials. Alternatively, incineration can be used with a large amount of energy to thermally decompose MSW (e.g., plastics), releasing energy, greenhouse gases, and other pollutants into the environment (Antelava et al., 2019; Devasahayam et al., 2019). MSW incineration may produce other byproducts, such as soot particles (e.g., PM2.5) and bottom ash residues that may act as carrier agents to transfer toxic chemicals to the environment. Although pollution control technologies reduce environmental impacts, incineration irreversibly converts plastics into other forms that are no longer usable from a closed-loop system standpoint (Oppelt, 1990; Quina et al., 2011; World Health Organization, 2001). Landfilling plastic EoL pathway, has been debated as both an impediment to further improvement in recycling and a necessity for storing nonrecoverable materials. Also, plastics may persist within the landfills for many years, creating an accumulation of solid waste and occupying land space (Zhang et al., 2021).

Chemical additive migration during the EoL plastic stage presents a concern for workers in the EoL plastic pathways, ecosystems, and the public. Regardless of the EoL pathway, the adverse impacts warrant a closer analysis of chemical additive movement to identify areas for improvement. Thus, this work aims to (1) complete a material flow analysis of plastics and chemical additives, (2) develop generic scenario analysis to estimate environmental releases and occupational exposure for risk assessment in day-to-day operation, (3) analyze the environmental burden, and (4) perform sensitivity analysis under different scenarios to justify changes to the existing EoL plastic management infrastructure. These analyses estimate the possible additive migration, greenhouse gas emissions, and other releases within the plastic life cycle and provide foundational knowledge to assist risk assessment efforts as mandated under the Toxic Substances Control Act (TSCA) and amended by the Frank R. Lautenberg Chemical Safety for the 21st Century Act (US EPA, 2022a, 2022b). Thus, stakeholders can evaluate potential risks and address any unreasonable risks chemicals may have on human health and the environment. Plastic management within the EoL pathways can also be optimized using the knowledge of chemical additive flow and potential chemical releases to design and alter the traditional linear production paradigm into a circular economy structure. In addition, identifying EoL pathways with high mass flow intensity can help prioritize the development of legislation and guidelines for improving existing and future EoL stages (Hernandez-Betancur et al., 2022, 2021a, 2021b).

2. Materials and methods

Creating a generic scenario for plastic management in the EoL stage requires a starting point for analysis. The United States MSW data published in 2018 was chosen to characterize material waste flow (US EPA, 2020a). Our analysis considers plastic movement and major types of discarded material collected. The mass flow intensity was then determined based on the magnitude of material movement. The chemical additive values reported in this work contain uncertainties because of the variation in product types. Plastic products require different amounts of chemical additives based on their intended application or condition of use.

The possible release and additive migration routes identified in this work are generally specific to collection, sorting, mechanical recycling, incineration, and landfilling stages. In each major EoL pathway, general information, such as the number of companies, facilities, and employees, was defined. However, there are inherent differences between the EoL pathways, which require separate analyses. Issues relating to plastics, chemical additive contamination, and material degradation ultimately affect the quality of the recycled products and create unwanted exposure to workers in the facility (Schyns and Shaver, 2020).

The environmental impact and sustainability of the current EoL plastics management were measured using the Sustainable Process Index (SPI). An SPI calculation for a given process provides a land area required to close the material loop and dissipate all emissions and wastes sustainably. The total calculated area comprises the area required to produce the raw material, provide energy, installation, and required staff area for the process, and accommodate products and byproducts (Narodoslawsky and Krotscheck, 1995). SPIonWeb, a web tool, was used to calculate the arable areas required for a given process (Narodoslawsky, 2015). The user defines the material inputs required to manufacture a particular product. This tool considers the history of the materials, releases, and impacts information in a database, then sums the contribution from all material inputs. The possible impact on the air, water, and soil can be estimated as an arable area. A low arable area correlates to a more sustainable process.

The numerical estimation associated with this work was completed with published data from the US EPA and key research articles (Crowl and Louvar, 2011; Ghosh and P, 2019; Hahladakis et al., 2018; Teuten et al., 2009; US EPA, 2020b, 2014a, 2014b; Wowkonowicz and Kijeńska, 2017). Our chemical additive tracking analysis occurs during the EoL pathways, from plastic waste collection to mechanical recycling, incineration, and landfilling. Also, all datasets and spreadsheets used to generate our results are available at https://github.com/USEPA/GS_End-of-Life_Plastic_Additives (Brinton et al., 2018; Crompton, 2007; Hahladakis et al., 2018; Horodytska et al., 2020; Intergovernmental Panel on Climate Change (IPCC), 2002; Jambeck et al., 2015; Ma et al., 2020; Teuten et al., 2009; United Nations, 2021; van Velzen et al., 2017).

3. Results and discussions

This section presents a material flow analysis of plastics and additives with emphasis on the EoL stage to assess the state of plastic management efforts within the United States. We then examined each major EoL stage in greater detail through the generic scenario analysis. This analysis estimated the releases of chemical additives and occupational exposure in day-to-day EoL plastic management operations within the United States. An environmental impact assessment was performed to estimate the theoretical burden that the existing EoL practice has on the environment and natural resource. Following the discussions on the state of EoL practices and hazardous releases in the United States, we performed sensitivity analyses under different scenarios to determine the effects of (1) altering plastic recycling rate, (2) using chemical recycling, and (3) including additive extraction post-mechanical recycling to the existing EoL plastic management infrastructure.

3.1. Material flow analysis of plastic life cycle

EoL plastic management was assessed using the MSW data in the United States in 2018. The composition of MSW in 2018 is shown in Fig. 2a, and the EoL plastic composition is shown in Fig. 2b. Papers, metals, glasses, and plastics are the primary components considered for recycling. The US EPA has estimated that over 35.7 million tons (32.4 billion kg) of plastic waste were generated in the United States in 2018. The municipal plastic waste is composed of 14.8% polyethylene terephthalate (PET), 17.7% high-density polyethylene (HDPE), 2.4% polyvinyl chloride (PVC), 24.1% low-density polyethylene (LDPE), 0.3% polylactic acid (PLA), 22.8% polypropylene (PP), 6.3% polystyrene (PS), and 11.7% other plastics. The EoL plastic categories coincide with polymer resin identification codes 1 through 7, with the addition of polylactic acid (PLA). Most of the recycling efforts have been allocated toward recovering PET, HDPE, LDPE, and a select group of uncategorized plastics. Up to 3 million tons (2.7 billion kg) (~8.4%) of the waste plastics were successfully recycled, 75.8% were landfilled, and 15.8% were incinerated (US EPA, 2020a). Recycled plastics are generally reprocessed into pellets to be used as raw materials for new plastics. Similar EoL pathways were reported by the United Nations, in which 9% of the global plastics were recycled, 79% were landfilled, and 12% were incinerated (Ghosh and P, 2019). This concerning fact suggested improving the existing plastics processing infrastructure to minimize excess environmental accumulation and toxic exposure. This information was used as the MSW stream composition for the material flow analysis.

Fig. 2.

(a) Municipal solid waste (MSW) composition in 2018. (b) Plastic waste composition in the United States in 2018. The overall recycling rate equates to 8.4%. The main plastic waste includes polyethylene terephthalate (PET), high-density polyethylene (HDPE), polyvinyl chloride (PVC), low-density polyethylene (LDPE), polylactic acid (PLA), polypropylene (PP), polystyrene (PS), and other uncategorized types.

Consumer plastics are commonly produced from a blend of polymer resins and chemical additives to achieve the desired characteristic for a particular application. The chemical additives may include, but are not limited to plasticizers (10–70 wt%), flame retardants (3–25 wt%), antioxidants (0.05–3 wt%), UV stabilizers (0.05–3 wt%), heat stabilizers (0.05–3 wt%), slip agents (0.1–3 wt%), lubricants (0.1–3 wt%), antistatics (0.1–1 wt%), curing agents (0.1–2 wt%), blowing agents (0.5–20.5 wt%), biocides (0.001–1 wt%), colorants (0.25–5 wt%), pigments (0.001–10 wt%), fillers (0–50 wt%), and reinforcements (15–30 wt%) (Carrott and Davidson, 1998; Hahladakis et al., 2018; Hummel, 2002; Scheirs, 2004). These composition ranges were included as part of the generic post-consumer plastic stream. An abbreviated list of chemical additives can be found in Table G1 of the Supplementary Information (SI). A more comprehensive additive list is available from Wiesinger et al., which identified over 2400 substances of concern used in plastics (Wiesinger et al., 2021).

The 2018 MSW plastic waste stream was subjected to a material flow analysis under the existing EoL infrastructure to assess the possible chemical additive migration, greenhouse gas emissions, leaching, degradation, and release of substances into the surrounding environment. Our analysis scope is limited to collection, sorting, and EoL pathways (mechanical recycling, incineration, and landfilling). Fig. 3 presents a high-level screening estimation of the chemical release within the Production, Use, and EoL stages of plastics using 2018 MSW data as the basis for calculation. The calculations done in this analysis were completed using assumptions tabulated in Table A1 in the SI, supported by published research (Brooks et al., 2018; Hahladakis et al., 2018; Law et al., 2020; Messenger, 2020; Ügdüler et al., 2020; Verma et al., 2016; Webb et al., 2012; Zhang et al., 2021). Table C1 in the SI provides the calculation results from the material flow analysis.

Fig. 3.

Material flow analysis of the plastic and additive mass flow using the 2018 municipal solid waste data as a basis. The black arrow indicates the main material movement, the red arrow indicates release as litter and spillage, the orange dashed line indicates migration and contamination, and the green arrow indicates recycling.

Plastic products are typically manufactured to possess high stability for long-term use, regardless of the actual application. Single-use plastics, for instance, persisted in the environment for several decades despite being made for one use. Following plastic disposal, the plastics are collected with other recyclable products like metals, glass, and paper. These recyclable materials are sent to a sorting facility. The allocation of plastic recycled, incinerated, and landfilled was determined by the US EPA waste characterization study (US EPA, 2020a). The plastic flow from the manufacturing stage through the EoL stage is not considered steady because plastic products can be reused and accumulate at various stages. However, tracking generic plastics throughout their life cycle requires a steady-state assumption. In addition, time is not factored into the calculation because this analysis aims to identify areas of concern during the plastic life cycle. Thus, the calculated values reported material inflow and outflow irrespective of the actual process duration.

A circular economy is not observed, given that only ~8.4% of plastic waste has been recycled. However, the reported recycling percent combines international plastic export intended for recycling (4.5%) and domestic recycling (3.9%) (Dell, 2019; Law et al., 2020; United Nations, 2021). The values associated with EoL plastic export and domestic recycling do not reflect the actual mass that was successfully recycled. Instead, the reported recycling values are mass that were sent out from the sorting facilities to domestic and overseas recycling facilities. State-of-the-art recycling techniques are not expected to achieve a 100% recovery rate. Additionally, Law et al. (2020) reported that 25 – 75% of US EoL plastic exports are mismanaged in the receiving country. The plastic wastes exported indirectly lead to additional additive releases into the environment because some receiving countries are not equipped to process the wastes and scraps.

Overall, the material flow analysis describes a high-level overview of plastic and additive mass movement throughout the plastic life cycle using the 2018 MSW data reported by the US EPA. The absolute values of calculated releases shown in Fig. 3 were transformed into a life cycle inventory, as summarized in Table 1, for estimating plastic and additive releases for any given mass flow basis (Hahladakis et al., 2018; Jambeck et al., 2015; Ritchie and Roser, 2018; US EPA, 2020a; Vanderreydt et al., 2021; Zheng and Suh, 2019). Although the release values are relatively small for a given mass unit input, processing billions of kg of plastic waste can raise the releases to an alarming level. Incineration and landfilling provide negligible mass output because these operations are designed to thermally degrade and contain solid waste, respectively. However, they do not solve the issue regarding resource circularity because the wastes are no longer recoverable. On the other hand, mechanical recycling does provide a positive mass output toward the manufacturing stage if we combine the treatment of plastic waste import and plastics sent for recycling from the sorting stage. In Fig. 4, the relative mass flow intensity of plastics and additives was summarized by considering the standard MSW management practices, efficiency issues, and plastic waste import and exports. As expected, we observed a heavy shift toward landfilling and incineration instead of mechanical recycling. More importantly, every EoL process that contains plastics has the potential to release chemical additives over time in less controlled environments (land, air, water, general contamination). In addition, the theoretical additive release routes can expand further to specific cases that include but are not limited to drinking water quality, hazardous substances in aquatic life, and recycling of other plastics. Thus far, the results are nationwide estimations based on the US 2018 MSW management. The uncertainty is further reduced in Sections 3.2–3.5, the generic scenario analysis of plastic waste and chemical additives. This analysis contains the numerical estimations of facility-level releases, contamination, and exposures under normal EoL operation.

Table 1.

Life Cycle Inventory of Plastics and Additives at the EoL Stages of the Plastic Life-Cycle.

End-of-Life Plastic Waste Management LCI
Materials	Input (kg/total kg input)	Output (kg/total kg input)	Releases to Land (kg/total kg input)	Releases to Air (kg/total kg input)	Releases to Water (kg/total kg input)	Greenhouse Gas Emissions (kg CO2-eq/kg input)
Collection and Sorting
PET	1.5E-01	1.4E-01	1.3E-02	Negligible	3.8E-04	4.1E-01
HDPE	1.7E-01	1.7E-01	8.0E-03	Negligible	2.3E-04	4.1E-01
PVC	2.3E-02	2.3E-02	0.0E+ 00	Negligible	0.0E+ 00	4.1E-01
LDPE	2.3E-01	2.3E-01	5.0E-03	Negligible	1.5E-04	4.1E-01
PLA	2.0E-03	2.0E-03	0.0E+ 00	Negligible	0.0E+ 00	4.1E-01
PP	2.2E-01	2.1E-01	0.0E+ 00	Negligible	1.0E-05	4.1E-01
PS	6.0E-02	6.0E-02	0.0E+ 00	Negligible	1.0E-05	4.1E-01
Other	9.0E-02	7.9E-02	1.1E-02	Negligible	3.4E-04	4.1E-01
Additives	5.6E-02	5.2E-02	4.0E-03	Negligible	1.2E-04	4.1E-01
Mechanical Recycling
PET	1.8E-01	1.9E-01	1.8E-05	4.7E-08	5.3E-07	−1.2E+ 00
HDPE	1.2E-01	1.2E-01	1.2E-05	3.0E-08	3.4E-07	−9.7E-01
PVC	1.3E-02	1.4E-02	1.3E-06	3.0E-09	4.0E-08	Negligible
LDPE	8.4E-02	8.7E-02	8.2E-06	2.1E-08	2.4E-07	Negligible
PLA	Negligible	Negligible	Negligible	Negligible	Negligible	Negligible
PP	9.0E-03	9.0E-03	8.7E-07	2.0E-09	3.0E-08	Negligible
PS	1.1E-02	1.2E-02	1.1E-06	3.0E-09	3.0E-08	Negligible
Other	3.7E-01	3.8E-01	3.6E-05	9.4E-08	1.1E-06	−1.1E+ 00
Additives	2.1E-01	1.8E-01	2.1E-05	5.4E-08	6.1E-07	−5.4E-01
Incineration
PET	1.4E-01	Negligible	1.4E-05	0.0E+ 00	4.0E-07	1.4E+ 00
HDPE	1.7E-01	Negligible	1.7E-05	0.0E+ 00	5.0E-07	1.4E+ 00
PVC	2.2E-02	Negligible	2.1E-06	0.0E+ 00	6.3E-08	7.4E-01
LDPE	2.4E-01	Negligible	2.3E-05	0.0E+ 00	7.0E-07	1.4E+ 00
PLA	2.0E-03	Negligible	2.0E-07	0.0E+ 00	6.0E-09	1.4E+ 00
PP	2.4E-01	Negligible	2.3E-05	0.0E+ 00	6.8E-07	1.4E+ 00
PS	6.5E-02	Negligible	6.3E-06	0.0E+ 00	1.9E-07	1.8E+ 00
Other	7.5E-02	Negligible	7.2E-06	0.0E+ 00	2.2E-07	2.6E+ 00
Additives	5.1E-02	Negligible	5.0E-06	1.5E-12	1.5E-07	1.1E+ 00
Landfilling
PET	1.3E-01	Negligible	1.9E-02	Negligible	5.6E-04	4.4E-02
HDPE	1.7E-01	Negligible	2.2E-02	Negligible	6.5E-04	4.4E-02
PVC	2.3E-02	Negligible	2.7E-03	Negligible	7.9E-05	4.4E-02
LDPE	2.4E-01	Negligible	3.0E-02	Negligible	8.9E-04	4.4E-02
PLA	2.0E-03	Negligible	2.4E-04	Negligible	7.1E-06	4.4E-02
PP	2.4E-01	Negligible	2.8E-02	Negligible	8.5E-04	4.4E-02
PS	6.7E-02	Negligible	7.9E-03	Negligible	2.4E-04	4.4E-02
Other	6.9E-02	Negligible	1.1E-02	Negligible	3.2E-04	4.4E-02
Additives	4.9E-02	Negligible	6.9E-03	Negligible	2.1E-04	4.4E-02

Fig. 4.

Sankey diagram to illustrate the mass flow intensity of the 2018 EoL plastic waste management in the US. All numbers should be multiplied by 10⁶ tons (US).

3.2. Generic scenario of plastic collection and sorting (I)

The current collection methods employed to gather post-consumer plastics are not expected to alter the quality of the substance considerably due to the short residence time between disposal collection and sorting (Luijsterburg and Goossens, 2014). Therefore, chemical additive transfer between materials can be neglected at the collection stage. However, plastic littering into the environment is a likely occurrence during material transfer and transport. For instance, a collection truck utilizing automation to lift and dump the content of the curbside blue recycling bin will not always successfully collect every material. Low-density materials may be swept out of the falling trajectory into the surrounding environment by wind or incorrect bin positioning before the collection process. Given the lack of spillage data during curbside collection, the US EPA recommends a spillage release rate of 0.01% (US EPA, 2014a).

Eq. (1) estimates that approximately 3570 tons (3.2 million kg) of plastics/yr. (R_{total,collection,spill}) can be spilled as litter during the collection stage. M_{plastic,collection} represents the mass rate of post-consumer plastics collected yearly and L_{collection,spill} is the loss fraction during collection due to spillage. The spilled plastics remain in the environment until they are manually removed. In some cases, the wind may sweep the plastics to a nearby wood, pond, or river, where the materials would reside for many decades and centuries. The chemical additives within the littered plastics are prone to migration to the environment over time due to partial degradation from UV and interactions with the surrounding medium (Sørensen et al., 2021). Wildlife may confuse the materials with food and consume the plastics unknowingly. Research has suggested that the digestive fluid within animals may act as an organic solvent to promote the leaching of chemical additives from plastics because traces of chemical additives accumulating within the tissues of the affected animals have been detected (Tanaka et al., 2015).

The US EPA has approximated that the fraction of chemical additives is 0.0005 – 0.70 kg/kg of plastic and 0.55 kg/kg of plastic on average (US EPA, 2014a). The expected range of chemical additive release during the entire lifetime of spilled plastics may thus range between 0.0005 and 0.70 kg/kg spilled. This estimation is time-dependent, which means the mass basis of 3570 tons (3.2 million kg) of plastics spilled per year can release between 1 and 2500 tons (907 – 2.3 million kg) of chemical additives to different sources during the plastic lifetime.

$${\text{R}}_{\text{total,}\text{collection,spill}}={\text{M}}_{\text{plastic,}\text{collection}}\cdot {\text{L}}_{\text{collection,spill}}$$ (1)

It should be noted that the litter generated during waste collection is a minor contribution to the current total litter fraction. Jambeck et al. (2015) has estimated that approximately 2% of all plastic waste generated globally becomes litter. Thus, the 35.7 million tons (32.4 billion kg) of plastics collected in 2018 may release up to 714,000 tons (648 million kg) of plastic, i.e., 350 – 500,000 tons (318,000 – 454 million kg) of associated chemical additives, across the entire plastic life cycle. Sorting methods may introduce chemical hazards to workers because manual sorting is required in some steps. Impurities and different material types may unintentionally be mixed in a specific pile, complicating downstream recycling processes. For example, in PET recycling, PVC has been a known contaminant capable of generating acids to degrade PET resin and subsequently degrade the mechanical and chemical properties of the polymer (Zhao, 2013). A concentration as low as 50 ppm PVC can cause considerable damage to the PET resins (Sommer, 1994; Zhao, 2013). Therefore, errors made during plastic sorting may lead to unintended complications during the reprocessing stage.

3.3. Generic scenario of plastic mechanical recycling (II)

This generic scenario analysis identified the major release points and potential release quantities during mechanical recycling. A high-level description of the mechanical recycling material release route is shown in Fig. 5a. Sorted plastics are typically sent to separate recycling facilities to process the specific material type. A typical practice of thermoplastic recovery includes particle size reduction, washing, rinsing, drying, and extrusion. Plastic wash has been reported to consume 2 – 3 m³ of water per ton of material (Hopewell et al., 2009). However, impurities from the consumer use stage and disposal can remain on the surface or within the plastic container even after a wash.

Fig. 5.

The major chemical additive release routes during (a) mechanical recycling, (b) incineration, and (c) landfilling. Solid lines are releases generated as a result of the process. The dotted line indicates releases prior to the process.

Additionally, chemical additives are present in the plastic from the manufacturing stage. More additives can be introduced into the recycling steps to ensure that the materials can be processed easily. The simplified illustrations shown in Fig. 5 are used to highlight the environmental release points, a crucial step for performing occupational risk assessment, rather than to demonstrate the technical detail of the process.

The generic scenario for mechanical recycling estimates the following:

• The number of mechanical recycling facilities in the US

• Releases to air, water, incineration, or landfill from material transport

• Releases during the extrusion process, including changing or cleaning of emission control filter (from dust and fugitive air emissions), equipment cleaning, and cooling water

• Number of workers that may come in contact with the chemical of interest during operation

• Inhalation and dermal exposures from material handling

• Release of chemical additives and volatile organic compounds to workers and nearby environment

• Chemical contamination incorporated into the reformed plastics after melt extrusion

3.3.1. General facility estimates

Material recycling facilities can be represented by NAICS Code 562920 (“NAICS Code 20 - Materials Recovery Facilities,”, 5629, 2020). There are 373 verified active companies in the United States with approximately 21,834 workers (“NAICS Code 20 - Materials Recovery Facilities,”, 5629, 2020). Each recycling facility across the US typically processes between 100 and 500 tons (90,719 – 453,600 kg) of plastics per day (Borchardt, 2000). The US EPA suggested a default estimate of an 8-hr workday and up to 250-days of operation per year for exposure estimation in the plastic converting industry (US EPA, 2014b). Material recycling utilizes similar processes as plastic compounding and converting industries. Therefore, the operation hours during mechanical recycling are assumed to be the same as the plastic compounding and conversion. The mass of plastics may contain between 0.05% and 70% of chemical additives and impurities from the previous use (Hahladakis et al., 2018). The general estimates presented thus far were used to analyze occupational exposure, releases, and contamination of unwanted substances during mechanical recycling.

3.3.2. Chemical additive use in mechanical recycling

Chemical additives, e.g., plasticizers, stabilizers, chain extenders, compatibilizers, and fillers, may be added throughout the mechanical recycling process to increase polymer processability. The amount added may range between 0.05% and 70% by weight (Schyns and Shaver, 2020; US EPA, 2014b). For instance, PET and HDPE have been blended with 5 – 15% compatibilizer before extrusion blow molding (Özcanli et al., 2002). A compatibilizer allows for the successful blending of two immiscible phases. Thus, the combined properties of the two polymers can overcome challenges related to recycling individual polymers if the waste materials would otherwise be discarded or considered acceptable for use in low-value applications (Schut, 2004). During mechanical processing, such as extrusion, chemical bonds may break in the main polymer chain, causing free radical formation (Özcanli et al., 2002; Schyns and Shaver, 2020). This issue is generally mitigated using thermal, light, and antioxidant stabilizers. Such addition is necessary because the polymers produced are stabilized with considerations for the use phase. However, these polymers can be exposed to oxygenated species, high temperatures, and UV light during the mechanical recycling phase, requiring more additive stabilizers (Schyns and Shaver, 2020). Stabilizers do not necessarily eliminate degradation and radical attack with 100% efficiency. Chain extenders have been reported as an inexpensive option to reverse the damage caused by degradation. Oligomers formed from polymer chain degradation would react with the chain extenders through its reactive site and effectively experience regrowth in the chain length. Fillers are an alternative to mitigate damages caused by chain degradation by increasing Young’s modulus, elongation at break, and impact strength while decreasing processability. Starch, cellulose, chitin, glass fibers, wood, and lignin are generally used for fillers (Schyns and Shaver, 2020). PVC recycling presents a challenge as the polymer becomes brittle in the EoL stage. Therefore, additional plasticizers are needed to increase the ductility during reprocessing. However, the most used plasticizers, phthalates, present a health safety hazard because they are susceptible to migration from within the polymer matrix to the surface over time. The differences in properties and structure between recycled and virgin plastics are governed by degradations that occur during conventional recycling processes. Additionally, while chemical additives are generally non-volatile solids and liquids, these substances may migrate to the polymer surface, contaminate equipment surfaces, and cause unintentional exposure. The potential chemical additives within plastics are listed in Section F of the Supporting Information (SI).

3.3.3. Releases and contamination in mechanical recycling of plastics

3.3.3.1. Plastics.

Plastics transferred during the recycling process are susceptible to spillage, which can be treated as mass loss to the environment. Plastic material processing may generate dust caused by abrasion of the plastics in granulated, cut, and pelletized forms (Plastic Industry Association and American Chemistry Council, 2017). The dust released from plastic processing is an inhalation hazard for workers without proper personal protective equipment (PPE). It serves as another source of plastic released into the environment without recovery. The US EPA recommends a conservative estimate for spillage and dust release as approximately 0.01% (US EPA, 2014b). Through Eq. (2), the possible release to water and land due to material spillage (R_mech,spill) is approximately 9 – 45.4 kg/(day⋅site). M_plastic,mech represents the plastics processed daily and L_mech,spill is the loss fraction due to spillage. Suppose a similar mass loss rate applies for all verified recycling facilities within the US. In that case, the conservative plastic release from mechanical recycling facilities due to spillage is 900 – 5000 tons/yr (816,500 – 4.5 million kg/yr). The released plastics can be similar to landfilled plastics because they are not recovered.

$${\text{R}}_{\text{mech,spill}}=\frac{{\text{M}}_{\text{plastic,mech}}}{{\text{N}}_{\text{site,mech}}}\cdot {\text{L}}_{\text{mech,spillage}}$$ (2)

3.3.3.2. Chemical additives.

The chemical additives used to improve recycled plastics processability have unintended consequences on human health and the environment. He et al. (2015) assessed volatile organic compounds (VOCs) from various plastic recycling workshops and identified the concentration of alkanes, alkenes, monoaromatics, oxygenated VOCs, chlorinated VOCs, and acrylonitrile emitted to the surrounding air during extrusion.

The total concentration of the emitted chemicals varies drastically between recycling facilities because of the difference in chemical composition and stability at the extrusion temperature (100 – 300 °C). The total indoor VOC concentration was estimated at various recycling facilities with a mean concentration of 3.9 mg/m³ for polystyrene, 2.1 mg/m³ for polyamide, 1.9 mg/m³ for polyvinyl chloride, 1.1 mg/m³ for polyethylene, and 0.60 mg/m³ for polycarbonates to the surrounding air. Monoaromatic compounds contributed 47.7–91.6% of the total VOC content across all recycling facilities. These hazardous compounds emitted from the extrusion processes may pose cancer risks to the lung, blood, brain, liver, kidney, and biliary tract. Non-cancer risks may include sensory, liver, kidney, and central nervous system damage. Exposure to chemical vapors such as benzene, toluene, ethylbenzene, styrene, methylene chloride, acrylonitrile, and trichloroethylene poses substantial cancer and non-cancer chemical risks (He et al., 2015). The recycling facility data analyzed in this study is based on poor ventilation and can be considered a worst-case scenario.

The chemical additives are susceptible to migration from the polymer matrix to the surface as non-volatile substances because these chemicals were mixed with the polymer during manufacturing rather than through a chemical reaction. Therefore, the connections between the chemical additives and the polymer matrix can be broken with a sufficient driving force. Tang et al. (2014) assessed the concentration of polybrominated diphenyl ethers (PBDEs), a flame retardant released into soils, sediments, and human hair. PBDE concentration around plastic mechanical recycling facilities was detected at 1.25 – 5504 ng/g (600 ng/g average) in soils, 18.2 – 9889 ng/g (1619 ng/g average) in sediments, and 1.50 – 861 ng/g (112 ng/g average) in human hair. PBDEs present a major environmental and health concern because these chemicals can persist in the environment and bioaccumulate upon exposure (Tang et al., 2014).

While polymer reprocessing is generally done in separate facilities to avoid plastic cross-contamination, this practice does not account for the unintentional transfer of chemicals to recycled plastics. Migrated chemical additives and degradation products formed during melt extrusion may adhere to the processing equipment (Höfer, 2012). Purging compounds are used to clean residual materials from the previous cycle. The equipment may be manually cleaned through brushing and scraping. The residual materials at the end of the equipment cleaning process present a possible source of chemical release to incineration or landfill. Eq. (3) describes the chemical release estimation during the equipment cleaning process. m_additive,_mech is the mass of chemicals added to improve plastic processability during recycling and L_{mech,cleaning} is the fraction of chemical loss from processed plastics. The US EPA recommends a 2% conservative estimate for chemical release rates during processing (US EPA, 2014a). For estimation purposes, the US EPA defaults the chemical additive fraction in plastic products ranges between 0.0005 and 0.70 kg chemical additive/kg plastic (US EPA, 2014a). Based on the typical plastic material flow of 90,719 – 453, 600 kg/(day⋅site), approximately 45.4 – 317,500 kg of chemical additives/(day⋅site) have been released from recycling equipment cleaning. Therefore, m_{additive,mech} exists to compensate for the mechanical property loss resulting from reprocessing and its value is assumed as 45.4 – 317,500 kg chemical additive/(day⋅site) added during the recycling process. The release due to equipment cleaning (R_{mech,cleaning}) is therefore 816.5 – 5.8 million kg/(day⋅site).

$${\text{R}}_{\text{mech,cleaning}}={\text{m}}_{\text{additive,}\text{mech}}\cdot {\text{L}}_{\text{mech,cleaning}}$$ (3)

The equipment cleaning operation is not expected to have 100% efficiency in removing the released chemical additives. Traces of contamination between batches have been reported from an analysis of recycled HDPE and LDPE (Horodytska et al., 2020). These plastics were found to contain approximately 30–100 μg contaminants/g plastic, 330–700 μg chemical additive degradation product/g plastic, and 30–800 μg chemical additives/g plastic (Horodytska et al., 2020). These substances may accumulate through every iteration of plastic recycling and cause unwanted exposure and release during the plastic use stage. Based on the level of unintentional substances found in recycled HPDE and LDPE plastics, we approximated the cleaning operation may possess an efficiency of up to 99.9% (Welle, 2008).

3.3.4. Occupational exposure in mechanical recycling

Workers in the EoL material management sector have experienced a higher illness rate relating to the respiratory system, including bronchitis and decreased lung function (Wouters et al., 2005). Handling plastics may generate airborne substances containing a mixture of dust, plastics, and volatile organic compounds released from material degradation. Our general facility estimate has shown that there are currently 373 verified active companies in the United States with approximately 21,834 employees (“NAICS Code 20 - Materials Recovery Facilities,”, 5629, 2020). An average of 58.5 workers per recycling facility can thus be assumed for calculation. This information does not account for the division of labor among the total workers in a given recycling facility. Not all workers can be exposed to the same level of inhalation hazard due to the variation in job function. However, workers at a recycling facility remain at a higher risk than the common public and should be included in a conservative estimate.

Inhalation exposure (EXP_{inhalation,mech}) can be represented by Eq. (4), where C_{particulate,mech} is the concentration of the particles of concern (15 mg/m³ for unregulated particulates based on TWA PEL)(National Institute for Occupational Safety and Health, 2011; OSHA, 2021), r_breathing is the rate of breathing (1.25 m³/hr)(US EPA, 2011), t_exposure is the duration of exposure (8 hrs/day), and F_{additive,plastic} is the fraction of additive in plastic (0.0005 – 0.70 kg/kg plastic and 0.55 kg additives/kg plastic average) (US EPA, 2014a). The inhalation exposure rate (EXP_{inhalation,mech}) for workers handling mechanical recycling can range between 0.075 and 105 mg particles/day. This estimation assumes insufficient PPE use, and thus the reported value is likely higher than the actual daily exposure. Assuming insufficient PPE usage provides a conservative estimate for designing a ventilation system to reduce the concentration of solid particles in the air.

$${\text{EXP}}_{\text{inhalation,mech}}={\text{C}}_{\text{particulate,mech}}\cdot {\text{r}}_{\text{breathing}}\cdot {\text{t}}_{\text{exposure}}\cdot {\text{F}}_{\text{additive,plastic}}$$ (4)

Direct contact of chemical additives with the skin during mechanical recycling is plausible because additives are typically added to ensure the materials can be processed efficiently. Given that chemical additives are generally solids, the potential dermal exposure was estimated using the EPA/OPPT Direct 2-Hand Dermal Contact with Solids Model (US EPA, 2014a). In Eq. (5), EXP_dermal,mech represents the potential dermal exposure to chemical additives per day, a_{incd,additive} is the mass of chemical additives in contact for a given incident (up to 3100 mg chemical additive/incident), N_{exp,incd,incn} is the number of exposure incidents per day (1 incident/day), and F_{additive,plastic} is the fraction of chemical additive in plastic (0.0005 – 0.70 kg/kg plastic and 0.55 kg chemical/kg plastic average). The dermal exposure rate for workers handling mechanical recycling can range between 1.6 and 2170 mg particles/day.

$${\text{EXP}}_{\text{dermal},\text{mech}}={\text{a}}_{\text{incd,additive}}\cdot {\text{N}}_{exp,\text{incd,mech}}\cdot {\text{F}}_{\text{additive,plastic}}$$ (5)

3.4. Generic scenario of plastic incineration (III)

Incineration is generally used to reduce the volume of waste materials while simultaneously recovering energy (Sahin and Kirim, 2018; US EPA, 2017). Plastic waste incineration involves the destruction of MSW through combustion and may reduce the methane gas generation from landfills (US EPA, 2020b). However, this process may negatively impact the environment if the pollutants from combustion are not controlled (Verma et al., 2016). Open burning of plastics may release vaporized dioxins, furans, mercury, and polychlorinated biphenyls into the atmosphere, creating persistent and unintentional hazards to human and animal health and the environment (Hopewell et al., 2009; Oppelt, 1990; Shaub, 1993; Verma et al., 2016).

The plastic incineration process begins with collected plastics unloading from the collection truck and storing them in a bunker (US EPA, 2020b). The plastics are then sent to combustion using a choice between a fluidized bed, grate technology, and rotary kiln (US EPA, 2020b). A fluidized bed combustor contains an emission control system, small boiler size, and generates no ash content. One possible configuration of a fluidized bed combustor includes using the operating temperature conditions at 450–550 °C, 64 bars, and 0.4 – 1.0 m/s of air or superheated steam. The polymer becomes volatile, leaving fibers, fillers, metals, and other non-volatile elements in the fluidized bed. A secondary combustion chamber is used at a higher temperature to ensure the complete oxidation of the plastics (Sahin and Kirim, 2018). Grate incineration is a one-stage combustion process operating at a temperature greater than 850 °C in the presence of oxygen. Bottom ash can be generated with this method (Jiang et al., 2020). Rotary kiln combustion operates between 900 and 1200 °C to incinerate waste, forming CO₂ and H₂O. A rotary kiln is advantageous over other incinerators because solid, liquid, and gas can be processed. The heat from the exhaust gas generated by the combustion of plastics gets transferred through a steam turbine to produce energy. Ash generated at the end of incineration is collected and transported to an enclosed building by leak-proof trucks for landfilling (US EPA, 2020b). The thermal destruction route of plastics is expected to release many substances that are not necessarily similar to the initial chemicals. Fig. 5b illustrates the major additive release routes described during incineration.

The generic scenario for plastic incineration aims to estimate the following:

• The number of MSW combustion facilities in the U.S (NAICS 562213) (“NAICS Code 13 - Solid Waste Combustors and Incinerators,”, 5622, 2022)

• Releases to air, water, or landfill from material transport

• Releases during incineration process, including changing or cleaning of emission control filter (from dust and fugitive air emissions), equipment cleaning, and cooling water

• Number of workers that may encounter the chemicals of concern during normal operation

• Inhalation and dermal exposures from gases generated during operation

• Release of chemical additives and volatile organic compounds to workers and nearby environment

3.4.1. General facility estimates

Waste incineration facilities can be represented by NAICS Code 562213 (“NAICS Code 13 - Solid Waste Combustors and Incinerators,”, 5622, 2022). In 2018, there were 61 verified active companies in the United States and approximately 2246 employees (Sahin and Kirim, 2018). A waste incineration unit may process between 50 and 1000 tons/day (45,360 – 907,185 kg/day) of materials. A typical incineration facility may include between two to three incinerators, leading to an approximate process flow rate between 100 and 3000 tons/day (90,719 – 2.7 million kg/day) for a given facility (Borchardt, 2000; Sorrels et al., 2017; Sustainable Sanitation and Water Management Toolbox, 2020). Based on the MSW data in 2018, approximately 12 – 360 tons of plastics/day (10,900 – 327,000 kg) are incinerated per facility. These facilities can operate continuously for up to 8000 hrs. (334 days) within a given year.

3.4.2. Releases from incineration

Like the previous cases, such as mechanical recycling, materials transported to incineration facilities are susceptible to spillage and thus can be treated as mass loss to the environment. The US EPA recommends a conservative estimate for spillage release of approximately 0.01% (US EPA, 2014b). Through Eq. (6), the possible release to water and land due to waste spillage (R_incn,spill) is approximately 1 – 33 kg plastics/(day⋅site). M_plastic,incn represents the total plastics processed by incineration in 2018. Incineration facilities are expected to handle a combination of different substances and plastics. Given that plastics account for approximately 12% of MSW, R_incn,spill only reflects the MSW plastic fraction. F_{incn,spillage} is the loss fraction due to spillage. If a similar mass loss rate applies for all incineration facilities within the US, the conservative yearly release from incineration facilities due to spillage is 24 – 733 tons (21,800 – 665,000 kg) of plastics/yr. The released plastics can be regarded similarly to landfilled plastics because they are not recovered.

$${\text{R}}_{\text{incn,spill}}=\frac{{\text{M}}_{\text{plastic,incn}}}{{\text{N}}_{\text{site,incn}}}\cdot {\text{L}}_{\text{incn,spill}}$$ (6)

The incineration process releases carbon dioxide and water vapor while producing byproducts such as nitrogen oxide, dioxins, furans, metals, acid gases, volatile chlorinated organic compounds, polycyclic aromatic compounds, and incombustible ashes. This process generally reduces the waste to 10 – 15% of the original volume and 20 – 35% of the initial mass as ash (Joseph et al., 2018). Emission control technologies such as electrostatic precipitators, fabric filters, wet scrubbers, and spray-dry absorbers can be used to remove and control airborne particles, hydrochloric acid, sulfur dioxide, dioxins, and heavy metals. Under the normal operating condition, the amount of pollutants released can be expected to be minimal because all emission control technologies are present to mitigate the hazards. However, a spike in pollutant release to the atmosphere may deviate beyond normal during incinerator startup and shutdown, after a change in waste composition, equipment malfunctioning, operator error, and poor incinerator maintenance (National Research Council, 2000). The incineration release estimates from this study utilize data during steady-state operations. Additional accuracy can be obtained with more emission data throughout the waste incineration process.

One ton (907 kg) of MSW incineration may generate up to 0.033 tons (30 kg) of air pollution control (APC) residues, 0.33 tons (300 kg) of bottom ash, and flue gas at a flow rate of 4500 – 6000 Nm³. Air pollution control (APC) residues are hazardous solid waste with high pH, volatile heavy metals, soluble salts, dioxins, and furans (Quina et al., 2011). In the absence of control devices, incineration may release flue gas up to 1 – 3% of the total ash content, containing inorganic ash, soot, and organic compounds deposited on fine particles (Joseph et al., 2018; National Research Council, 2000). The general composition of the pollutants and the permissible release concentration are shown in Table B1 in the SI. The pollutants exit a typical incineration furnace at high concentrations. Thus, a gas cleaning system is generally implemented under normal conditions to remove fly ash, acids, dioxins, and furans before the gas exits the flue-gas stack (Quina et al., 2011). The desired concentration can be achieved using a combination of wet scrubbers, dry multi-cyclone, semi-dry scrubbers, selective catalytic reduction, electrostatic precipitator, fabric hose filter, and activated carbon (European Commission. Joint Research Centre, 2019).

Given that plastic wastes are commonly incinerated with other MSWs, the release contribution from plastic incineration must be estimated separately to assess the environmental impacts of the thermal destruction of plastics. PVC, PTFE, and plastics containing brominated flame retardants have been reported to release flue gas containing a considerable level of halogens, acid gases, and persistent organic pollutants such as dioxins and furans compared to the remaining plastics (Hahladakis et al., 2018; Johnsen, 2015). In 2018, over 5.6 million tons (5.1 billion kg) of plastics (15.8% of total MSW) were sent to incineration facilities, with 2.6% (1.7 billion kg) of the incinerated plastic being PVC (Jiang et al., 2020; Johnsen, 2015; US EPA, 2020b). Weber and Kuch have shown that considerable formation of dioxins, furans, and other potentially toxic substances can occur due to incomplete combustion of plastics (Weber, 2003). PVC is considered the highest contributor to toxic substance release because it provides up to 38 – 66% chlorine content in MSW. Brominated aromatic flame retardants may contribute to brominated compounds concentration in flue gas and ultimately support the synthesis of dioxins and furans (Weber, 2003). Incineration experiments have shown that brominated-chlorinated dibenzodioxins and dibenzofurans concentration in uncontrolled waste burning can range between 1 and 9000 μg/g solid combustion residue and 0.8 – 1700 μg/m³ flue gas (Weber, 2003). Many of these chemical formations are influenced by the incomplete combustion of brominated flame retardants (BFRs) in plastics. Controlled combustion (temperature >850 °C) may be used to thermally degrade between 90% and 99.9% of the brominated flame retardants, thus inhibiting the formation of furans and dioxins (de Wit, 2002; Weber, 2003). The Clean Air Act of 1970 has prohibited incineration facilities from practicing uncontrolled burning to ensure that emission regulations are met (US EPA, 2020b). A conservative estimate can be made by treating the release factor of BFR to be representative of other chemical additive release from incinerating MSW plastics.

Based on the reported mass of incinerated plastic in 2018 (US EPA, 2020a), the estimated chemical additive release from incineration may range between 4546 – 408,200 kg/(yr⋅site) that end up in the incinerator ash as a result of incomplete combustion. Incinerator bottom ash may include the remaining MSW that can no longer be burned. Additional treatment can remove valuable metals from the bulk ash. The remaining ash can be used in construction (Joseph et al., 2018). The chemical additives released as fly ash are subjected to air pollution control technologies to ensure that emission regulations are met, according to the requirements listed in Table A1. Unmitigated fly ash may release chemical additives into the atmosphere at a rate of 136 – 12, 250 kg/(yr⋅facility) (Quina et al., 2011). The removal efficiency of 99.9% may reduce this release rate to 0.14 – 12.2 kg/(yr⋅facility).

Dispersion modeling was used to estimate the potential exposure from incineration flue gas. The scenario of release from an incineration facility can be represented by the Pasquill-Gifford model for a plume with a continuous steady-state source at height above ground level with wind moving in the x-direction at constant velocity (Crowl and Louvar, 2011). Eq. (7) displays the model for predicting the average flue gas concentration $(\overline{\text{C}})$ at a specified coordinate from the source. The x-direction represents the direction of the wind, y is the crosswind, and z is the vertical direction. The concentration is a function of the gas flow rate (Q_m), average gas velocity $(\overline{u})$, dispersion coefficient (σ) in the x, y, and z directions, and the stack height (H_r) (Crowl and Louvar, 2011).

$$\begin{gathered} \overline{\text{C}}(\text{x},\text{y},\text{z})=\frac{{\text{Q}}_{\text{m}}}{2\pi {\sigma }_{\text{y}}{\sigma }_{\text{z}}\overline{\text{u}}}exp\left[-\frac{1}{2}{\left(\frac{\text{y}}{{\sigma }_{\text{y}}}\right)}^2\right]\cdot \{exp\left[-\frac{1}{2}{\left(\frac{\text{z}-{\text{H}}_{\text{r}}}{{\sigma }_{\text{z}}}\right)}^2\right] \\ +exp\left[-\frac{1}{2}{\left(\frac{\text{z}+{\text{H}}_{\text{r}}}{{\sigma }_{\text{z}}}\right)}^2\right]\} \end{gathered}$$ (7)

The daytime maximum ground-level concentration in the wind direction is 0.39 – 0.69 ppm at approximately 23 – 322 m (0.014 – 0.200 mi) from the source for an urban environment. Alternatively, the daytime maximum ground-level concentration in the wind direction in rural settings is 0.12 – 0.28 ppm at approximately 55.5 – 1203 m (0.034 – 0.748 mi) from the source for an urban environment. Table B2 in the SI summarizes the parameters used for estimating the flue gas dispersion (“Annual Average Wind Speed in US Cities, ”, 2021; “Average annual temperature in the U.S. from, to, ”, 1895, 2020, 2021; “Barometric Pressure Summary, ”, 2021; Crowl and Louvar, 2011). Dow Chemical and the US EPA have tested the effect of transient combustion and established that particulate matter concentration increases by 100% from steady-state combustion (National Research Council, 2000). Transient state combustion is likely to occur during startup and shutdown. However, such an operation is not common throughout a given year because incineration can be continuously operated with minimal downtime.

Pollutants released from the incinerator may be dispersed into the environment, exposing the population to potentially toxic substances through inhalation. The pollutants may contaminate and persist in the nearby waters and food supply. If the pollutant concentrations exceed the permissible exposure limits, humans and animals consuming the contaminated substance may be subjected to short- and long-term side effects. Workers at an incineration facility are at a higher risk for exposure to toxic chemicals than the general public living in the surrounding area (National Research Council, 2000). Without proper PPE, toxic chemical exposure through inhalation and skin contact may become possible during solid waste feed handling and incinerator cleaning (Federal Remediation Technologies Roundtable, 2020). Chemical additives and other degradation products may contaminate solid ash post-incineration. Maintenance workers entering the incinerator to clean, inspect, or repair are highly susceptible to exposure to residual solid waste. These wastes result from incomplete combustion products, including polychlorinated biphenyls (PCB), polychlorinated dibenzofurans (PCDF), dioxins, and carbon monoxide (Stanmore, 2004). Daily cleaning is generally performed around the incinerator grate and ash pit to prevent clogging.

3.4.3. Occupational exposure in mechanical recycling

Inhalation exposure in incineration can be represented by Eq. (8), where C_{particulate,incn} is the concentration of the particles of concern (15 mg/m³ for unregulated particulates based on TWA PEL)(National Institute for Occupational Safety and Health, 2011; OSHA, 2021), R_breathing is the rate of breathing (1.25 m³/hr)(US EPA, 2011), t_exposure is the duration of exposure (8 hrs/day), and F_additive,ash is the fraction of chemical additives in incinerator ash (0.000015 – 0.0014 kg chemical additives/kg ash, as estimated from additive released and ash generated in incineration). Thus, the inhalation exposure rate (EXP_{inhalation,incn}) for workers in an incineration facility can range between 0.0023 and 0.21 mg chemical additives/day.

$$EX{P}_{\text{inhalation,incn}}=C_{\text{purticulate,}\text{incn}}\cdot R_{\text{breathing}}\cdot t_{\text{exposure}}\cdot F_{\text{additive,}\text{ash}}$$ (8)

The EPA/OPPT Direct 2-Hand Dermal Contact with Solids Model was used to estimate potential dermal exposure in mechanical recycling (US EPA, 2014a). In Eq. (9), EXP_dermal,incn represents the potential dermal exposure to chemical additives during the incineration process (cleaning ashes from the incinerator) per day, a_incd,ash is the mass of ash in contact for a given incident (assumed 5000 mg ash/incident)(Tyson, 2017), N_{exp,incd,incn} is the number of exposure incidents per day (1 incident/day), and F_additive,ash is the fraction of chemical additives in ash (0.000015 – 0.0014 kg chemical additives/kg ash, as estimated from chemical additives released and plastics processed in incineration). The dermal exposure rate (EXP_dermal,incn) for workers handling incinerator ashes can range between 0.075 and 7 mg chemical additives/day.

$${\text{EXP}}_{\text{dermal,}\text{incn}}={\text{a}}_{\text{incd,}\text{ash}}\cdot {\text{N}}_{exp,\text{incd,incn}}\cdot {\text{F}}_{\text{additive,}\text{ash}}$$ (9)

3.5. Generic scenario of plastic landfilling (IV)

Landfills are designed as solid waste containments, as an option regulated under the Resource Conservation and Recovery Act (RCRA), subtitle D (solid waste), and subtitle C (hazardous waste) (US EPA, 2020c). The types of landfills may include those intended for municipal solid waste (MSWLFs), industrial waste (construction and demolition (C&D) debris and coal combustion residual), hazardous waste, and polychlorinated biphenyl (PCB). MSW landfills, specifically, can generate and release an excess of methane gas, capable of trapping heat in the atmosphere with 28 – 36 times more potency than carbon dioxide. A gas collection system generally collects methane gas with 38 – 88% efficiency. This action reduces the carbon emissions from landfills, air pollution, and energy cost (US EPA, 2017).

Landfills are built with several protective measures to prevent unwanted substances from leaching into groundwater. The content of groundwater is regularly monitored to ensure that waste materials do not leach into the groundwater supply. A composite liner establishes the capacity of a landfill and blocks unwanted substances from leaching into the groundwater supply using a combination of clay and plastic liner. A collection pipe is used on the plastic liner to collect leachates for treatment. A crushed rock layer is placed around the collection pipe to prevent waste materials from clogging the pipe. A gas collection system is implemented to collect and store landfill gas (50% methane and 50% carbon dioxide) for future use. Solid wastes are dumped above the crushed rock layer and then covered using soil to reduce the odor. Landfills that have reached the maximum capacity for trash must have a final cap consisting of synthetic plastic and a layer of dirt over the landfill. Plants are grown above the plastic cap to reduce the chance of erosion (Thorneloe et al., 2020). Fencing is used as a last line of defense against wind-blown waste litter. However, landfilling operation is a source for unintentional chemical releases, as shown in Fig. 5c.

The generic scenario for plastic landfilling aims to estimate the following:

• The number of landfills in the US

• Number of workers that may come in contact with potential toxins

• Releases to air and water (methane gas, plastic leak)

• Release from waste transport

• Inhalation and dermal exposures

3.5.1. General landfill estimates

Solid waste landfills can be represented by NAICS Code 562212 (“NAICS Code 12 - Solid Waste Landfill,”, 5622, 2020). There are currently 651 verified active companies in the United States with 20,786 employees. The US EPA has an estimated 2627 active landfills in the United States as of March 2021 (“NAICS Code 12 - Solid Waste Landfill,”, 5622, 2020). A landfill site may follow different hours of operation. Once a sufficient amount of waste has been stored, landfills continue to generate leachate and gas to the surrounding environment regardless of the operation status. Therefore, a simplification was made, which assumes 365 days of operation to calculate releases. Based on the inflow of waste in 2018, an average landfill may receive over 55,000 tons of waste/day (50 million kg/day). However, landfill sites are not equivalent in capacity across the United States. The actual values associated with this vary greatly depending on the population density and geographic location.

3.5.2. Releases in landfilling

Despite the multitude of safeguards incorporated into landfills, littering is a common problem that can release waste to unexpected areas, including homes, roads, water supplies, and the ocean. Waste may be blown out of the vehicle during waste transport to the landfill and create litter in the street. The wind remains an essential factor during the waste dumping process. A semi-permanent litter fencing can be used to mitigate a portion of the plastic litter swept by daily unloading operations. Landfill gas may escape the gas collection system and into the atmosphere through cracks and leaks around the containment (Thorneloe et al., 2020).

We have previously estimated that 714,000 tons (648 million kg) (R_{total,litter)} of plastic waste generated became litter throughout the EoL processing for a given year. To maintain the material balance, the releases of plastic mass have been compiled and used to calculate the rate of plastic release from landfills as litter. R_mech,spill may range between 900 and 5000 tons/yr (816,500 – 4.5 million kg/yr), R_{collection,spill} equates to 3570 tons/yr (3.2 million kg/yr)., and R_incn,spill is 24 – 733 tons/yr (22,000 – 665,000 kg/yr).

Eq. (10) estimates that approximately 620,000 – 670,000 tons plastics/yr (562 million – 608 million kg). (R_{total,landfill,spill}) can be spilled as litter during the landfilling operation based on a 35.7 million tons (32.4 billion kg) plastics basis. The landfill contribution toward chemical additives release can thus reach 310 – 470,000 tons (281,200 – 426.4 million kg) over the lifetime of plastics, based on the typical mass fraction of chemical additives in plastics (0.0005 – 0.70).

$$R_{\text{total,}\text{landfill,spill}}=R_{\text{total,}\text{itter}}-R_{\text{total,}\text{collection,spill}}-R_{\text{total,}\text{mech,spill}}-R_{\text{total,}\text{incn,spill}}$$ (10)

Rainwater and groundwater may penetrate through the containment and into the landfill containing mixed MSW, presenting a complex environment that may cause chemical additives to be released from the polymer matrix (Wowkonowicz and Kijeńska, 2017). The contaminated rainwater acts as leachate, containing organic compounds resulting from plastics and bacteria activity throughout the degradation process. Landfill sites in industrialized countries have been known to perform leachate treatment, such as using aerobic and membrane bioreactors to reduce the BPA concentration to 0.11 – 30 μg/L (Asakura et al., 2004; Teuten et al., 2009). Without proper leachate treatment, chemical additives and BPA could be released into the environment and contaminate the nearby water supply. The rate of leachate release has been estimated to vary between 20% and 30% of the waste in the landfill (Youcai, 2018). Our generic scenario analysis holds that over 146 million tons (132.4 billion kg) of waste have been sent for landfilling in 2018, with each landfill receiving, on average, 55,000 tons (50 million kg) of MSW/day. The potential leachate generated from landfills may approach 10 million – 15 million kg/(yr⋅site). For a given site, the estimated yearly chemical additive release through leachate equates to 100 – 150 kg/(yr⋅site) (0.001% additive in leachate) (Teuten et al., 2009).

3.5.3. Occupational exposure in landfilling

The transmission of chemical additives to humans due to landfill activity is not straightforward. Plastics, chemical additives, and the corresponding degradation byproducts leave the containment as leachate or litter. Animal species such as birds, fish, and other marine mammals are likely to ingest these materials by mistake because of their similarity to food. The contaminants have been shown to desorb from plastics in gastric conditions and expose the organism to unintended toxicity (Teuten et al., 2009). Humans or other animals that consume the affected organisms can indirectly expose themselves to the contaminants. The concentration varies depending on geographic location and the rate of leachate and litter release into the environment. The resultant hazardous effects of chemical additives can vary depending on the organism type, duration of exposure, and chemical interaction within the biological system (Sridharan et al., 2022). Additionally, the chemical additive desorption rate from plastic depends on the size, type, and state of the plastic. For example, rubbery plastic items do not have many adsorption sites for chemicals; however, glassy plastic items possess high rigidity and tiny pores that serve as adsorption sites.

Workers at landfill sites can be directly exposed to potentially hazardous substances primarily through dermal contact. Sharp objects may temporarily injure workers and expose them to toxins that cause severe health complications, including disease, infection, and cuts. Liquid leachate may also be absorbed through the skin, posing more potential hazardous issues. Like other EoL processing methods, not all workers are exposed to the same level of dermal hazard because their job function varies. However, the likelihood of exposure for workers in the waste management industry is much higher than the common public.

Chemical additives in leachate are present at approximately 0.00001 kg of chemical additives/kg of leachate. Landfill workers are not expected to be exposed to chemical additives through the inhalation route under normal conditions. However, other volatile substances such as methane gas generated from MSW degradation remain constant.

Direct contact with chemical additives may occur during landfilling operations because there may be instances requiring workers to handle waste materials directly. Accumulated leachate in the landfill sites may contain a small fraction containing dissolved chemical additives. The potential dermal exposure was estimated using the EPA/OPPT Direct 2-Hand Dermal Contact with Liquid Model (US EPA, 2014a). EXP_{dermal_landfill} represents the potential dermal exposure to chemical additives in landfilling per day, b_{liq,skin,landfill} is the amount of liquid remaining on the skin (0.7–2.1 mg chemical/(cm²_⋅incident), A_surface is the surface area of contact (1070 cm² for two hands), N_{exp,incd,landfill} is the number of exposure incidents per day (1 incident/day), and F_{additive,leachate} is the fraction of chemical additive in leachate (0.00001 kg chemical additives/kg leachate). Through Eq. (11), dermal exposure rate (EXP_{dermal,landfill}) for landfill workers can range between 0.007 and 0.02 mg additives/day.

$${\text{EXP}}_{\text{dermal,}\text{landfill}}={\text{b}}_{\text{liq,skin,landfill}}\cdot {\text{A}}_{\text{surface}}\cdot {\text{N}}_{exp,\text{incd,landfill}}\cdot {\text{F}}_{\text{additive,}\text{eachate}}$$ (11)

The generic scenario calculations were categorized as the following: plastic wastes processed, plastic spilled/released, chemical additives released, inhalation exposure, and dermal exposure. There is limited data on chemical additive release in the collection and sorting stages. Therefore, the estimation for this stage was made from a holistic standpoint. Recycling, incineration, and landfilling EoL routes were estimated on a mass per facility per given year. It is expected that the number of EoL facilities, workers, recycling, incineration, and landfilling rates may change as a function of time. However, these generic scenario values can assist with identifying challenges regarding plastic recycling and the danger of toxic chemical release in typical operations.

3.6. Environmental impact assessment

The current methods of plastic waste management release materials into the environment and expose workers in the EoL stages, creating unintended and long-lasting consequences to health and the ecosystem. In Sections 3.1–3.5, we have quantified the potential plastics and additives mass flow throughout the EoL stages and estimated generic facility-level releases to highlight additive exposure risks. This section estimates the environmental consequences of the existing EoL practices by assessing energy footprints, greenhouse gas (GHG) releases, and ecological footprints to identify the environmental burden and motivate the shift toward a circular economy approach with minimal releases.

3.6.1. Releases

The GHG releases associated with plastics throughout the plastic life cycle have been calculated using the EPA Waste Reduction Model (WARM) (US EPA, 2015). The calculated mass flow and composition of the generic plastic waste stream were multiplied by the GHG emission factors shown in Table D1 (US EPA, 2015). Through incineration, up to 8.6 million tons (7.8 billion kg) of CO₂-eq in 2018 were generated from plastic components. Up to 1.2 million tons (1.1 billion kg) of CO₂-eq can be generated annually from landfilling plastics. By choosing mechanical recycling for 8.4% of the wasted plastic, up to 2.9 million tons (2.6 billion kg) of CO₂-eq emission can be prevented. If the United States can double the recycling rate to 16.8%, over 5.8 million tons (5.3 billion kg) of CO₂-eq emissions can be prevented. As shown in Fig. 3, up to 648 million kg (2%) of plastic waste generated can be released into the environment for a given year because of littering in the EoL stage. Most littering comes from landfilling operations because it is designed to store material waste. A greater volume of materials is expected to enter landfill sites than a recycling or incineration facility. The released plastics from littering equate to approximately 31,500 tons (28.6 million kg) CO₂-eq of emissions. While the magnitude of overall plastic litter is far lower than direct landfilling, littering presents a source of uncontrolled release. Each year, more materials may become litter and increase the impacts accordingly.

The world produces more than 360 million tons (326.6 billion kg) of plastics annually from a global perspective (Ritchie and Roser, 2018). Assuming the cumulative plastic waste generated is subjected to the United States plastic recycling infrastructure, there is a potential annual chemical additive release of 1.25 million tons (1.1 billion kg) and a GHG release of 111 million tons (100 billion kg) of CO₂-eq. However, many countries worldwide do not necessarily possess state-of-the-art recycling technologies to minimize toxic chemical releases, creating large uncertainties in the release estimations. Although analyzing the global EoL plastic management is beyond the scope of our work, efforts to improve EoL management processes are needed to achieve a circular economy.

3.6.2. Sustainable process index

The Sustainable Process Index analysis was used to assess the ecological footprint of EoL plastic management for 2018. SPI calculates the total arable area required to sustainably achieve one unit of the desired product for a given process. In EoL plastic management, the successful recycling of 1 kg of waste plastics equates to 371 m².a, where ‘a^′ signifies arable (Narodoslawsky and Krotscheck, 1995). The processes and the required arable areas to obtain 1 kg of recycled plastic have been listed in Table 2.

Table 2.

Arable area contribution required to obtain 1 kg of plastic materials from post-consumer plastic waste.

Processes	Inventory	Arable Area (m2. a/kg)	% Total Area
Main Product
Recycled Plastics	1 kg	371	100
Sub-Processes
Waste plastics mixture	11.8 kg	0.261	0.1
Municipal waste collection by 21 metric ton lorry	1 tkm	0.297	0.1
Treatment of waste plastic, mixture	2.17 kg	187	50.3
Extrusion (Final processing)	1 kg	144	38.9
Treatment of MSW, incineration	4.03 kg	39.8	10.7
Waste in a landfill for inert matter	19.4 kg	0.0230	0.0

The required area for recycling 1 kg of plastic ultimately impacts fossil carbon, water, air, soil, renewable, land area, and non-renewable resources. Based on the material flow in 2018, the management of plastics consumed water for 226 m².a/kg, fossil carbon equivalent to over 116 m².a/kg, soil for 16 m².a/kg, and air for 13 m².a/kg. Figure F1 in the SI illustrates this distribution of the resource impacts. The footprint impacts on renewable, non-renewable resources and land area are negligible and thus were excluded.

The current land surface area of Earth equates to 149 trillion m² (57.51 million mi²) (The Earth, 2021). The arable land was estimated to be 10.83% of the total land surface area (16.1 trillion m²) in 2018 (Food and Agriculture Organization, 2021). The United States could recycle up to 3.02 million tons (2.7 billion kg) of post-consumer plastics. The SPI analysis from disposal to recovery indicated that 1.02 trillion m².a is required to embed this process sustainably. The plastic management impact on the land is equivalent to 7% of the total arable land area on Earth. It is crucial to note that SPI serves as a theoretical representation of sustainability and is standardized as arable land area. Over time, the arable land area on Earth may change depending on land use, pollution, and erosion. In the present state, the United States EoL plastic management process is not sustainable if we include other post-consumer wastes such as food, glasses, metals, papers, and more in the analysis. Although the Earth does replenish resources over time due to material cycles, the rate of resource gain is drastically lower than the resource loss. Responsible allocation of spent resources ensures the annual carbon consumption is minimized.

The environmental footprint and release rates calculated in this work were based on the US EPA 2018 data on plastic waste management. Although the parameters used to estimate plastics and chemical additives releases are expected to change between years, a sensitivity analysis study can be used to estimate the trend of releases and impacts.

3.7. Sensitivity analysis

We analyzed three distinct scenarios to predict the impacts of (1) altering the efficiency of existing EoL plastic management, (2) incorporating chemical recycling techniques as a primary plastic management route, and (3) extracting chemical additives from bulk recycled plastics before reuse, on chemical additive release, GHG emissions, and energy footprints. The emission factors and energy footprints used in this study are tabulated in Table D1 in the SI (Devasahayam et al., 2019; Jeswani et al., 2021; REMADE INSTITUTE, 2019; Smeaton, 2021; US EPA, 2015; Vollmer et al., 2020).

3.7.1. Scenario 1: recycling efficiency is increased beyond the current rate to a maximum technical feasibility point

This hypothetical scenario predicts the effects of increasing the mechanical recycling rate on the global warming potentials, chemical additive releases, and energy footprint (REMADE INSTITUTE, 2019). The maximum technical feasibility of plastic recovery from the collection, sorting, and mechanical recycling can theoretically reach a maximum recovery of 72% (Brouwer et al., 2020). Therefore, the recovery rate of the plastics sent for recycling could theoretically be between 0% and 72%. Plastic waste export value was held constant at 4.5% regardless of the increase in recycling efficiency (Zhao et al., 2021). Incineration and landfilling are selected as the secondary method for processing non-recyclable plastic and were held at a constant ratio of 17.2:82.8. Fig. 6a shows that the increase in mechanical recycling rate effectively increases the total chemical additive release, greenhouse gas emissions, and energy footprint.

Fig. 6.

Sensitivity analyses of three hypothetical scenarios and their impacts on the total chemical additive release, greenhouse gas emissions, and energy footprint. (a) Scenario 1 examines the effect of increasing the rate of plastic recycling to the maximum technical feasibility limit; (b) Scenario 2 implements pyrolysis as a secondary plastic waste processing method, in addition to mechanical recycling; (c) Scenario 3 includes a chemical additive extraction stage before recycled plastics are sent to the manufacturing stage. The total plastic additive release has been redefined as total plastic additive removed to reflect on the purpose of the additive extraction process.

The increase in the chemical additive release is proportional to the chemical additive contamination in recycled plastics. Unrecyclable plastics are sent to incineration and landfilling, creating opportunities to release harmful gas-phase emissions and accumulation of plastic mass, respectively (Asakura et al., 2004; Federal Remediation Technologies Roundtable, 2020; Sustainable Sanitation and Water Management Toolbox, 2020; World Health Organization, 2001). These estimations were highlighted in our generic scenario analysis. Chemical additives may be released slowly into the surrounding area as leachate over time. Mechanical recycling is expected to release chemical additives at a higher rate than the other plastic waste processing methods within a given period. However, this statement should not detract from the merit of increasing the plastic recycling rate. The successful recovery of EoL plastic reduces the potential release and accumulation of plastics in the environment. The chemical additives would be accumulated in a more controlled environment within the recycled plastics. Additional processing may be implemented to reduce harmful chemicals from recycled plastics. Conversely, non-recycled plastics that accumulate in the environment can release harmful chemical additives to the ecosystem without control. Fig.C1 in the SI summarizes scenario 1 by illustrating the overall movement of plastics, chemical additives, and other MSW at the EoL stages based on 2018 MSW data on plastics.

3.7.2. Scenario 2: chemical recycling is used in conjunction with mechanical recycling

This hypothetical scenario examines the effects of implementing pyrolysis to treat all plastics that were not successfully recycled through the mechanical route and the associated global warming potentials, chemical additive releases, and energy footprint. Mechanical recycling remains the primary chosen method for plastic recycling. All untreatable plastic waste and solid residues resulting from mechanical and chemical recycling are sent to incineration and landfilling at a constant ratio of 17.2:82.8. International plastic waste export remains constant at 4.5%.

Pyrolysis was chosen to represent chemical recycling because it has been used commercially with a conversion efficiency range between 60% and 95% (Solis and Silveira, 2020). Alternative other chemical recycling techniques may include catalytic cracking, conventional gasification, plasma gasification, depolymerization, and hydrocracking. The life cycle inventory for the pyrolysis process was estimated based on the values reported by (Jeswani et al., 2021). The mechanical recycling efficiency was constant at 66.7% (Brouwer et al., 2020). The pyrolysis conversion efficiency was held at 95% (Solis and Silveira, 2020). Fig. 6b shows the increase in chemical recycling (pyrolysis) rate decreases the total chemical additive release, GHG emissions, and energy footprint. This reduction is expected because pyrolysis chemically converts plastics into energy, gas, oil, and solid residues (Anuar Sharuddin et al., 2016; Jeswani et al., 2021; Solis and Silveira, 2020). Other techniques, such as depolymerization, can recover the original monomer and thus provide opportunities for upcycling (Han, 2019). Harmful chemicals are converted into new products without damaging the identity of the materials. Therefore, the combination of mechanical and chemical recycling is a feasible approach to processing and minimizing the releases of plastic wastes. Figure C2 in the SI summarizes scenario 2 by illustrating the overall movement of plastics, chemical additives, and other MSW at the EoL stages with considerations for chemical recycling.

3.7.3. Scenario 3: chemical additives are extracted from mechanically recycled plastics

This hypothetical scenario examines the effects of implementing an extraction technique post-mechanical recycling on global warming potentials, chemical additive releases, and energy footprint. Chemical additive extraction is promising when performed as a solid-liquid extraction with dissolution-precipitation (Ügdüler et al., 2020). Common extraction types may include shake-flask extraction, Soxhlet extraction, ultrasonic extraction, microwave-assisted extraction, supercritical fluid extraction, accelerated solvent extraction, and dissolution-precipitation (Ügdüler et al., 2020). The success rate of these methods is highly dependent on the additives, plastics, and extraction conditions. The authors identified that dissolution-precipitation is preferable in the case of a broad range of chemical additives. However, chemical additives may degrade because of the high-temperature operation. Chemical additive degradation is not essential for this scenario because the extracted chemicals are present as a mixture in small quantities. Additionally, the separation and purification of the individual chemical additive components are not expected to be economically viable because of the large variety of substances present. Therefore, the extracted chemical additives were treated as process waste. While there may be economic value to the chemical additives extracted because these chemicals are generally organic compounds, this analysis is beyond the scope of our work.

Dissolution-precipitation showed that CO₂ savings equate to 65–75 wt%, as opposed to incineration and landfilling, because there is no chemical bond destruction (Vollmer et al., 2020). Therefore, the global warming potential for this process has been estimated to be approximately 30% of the incineration of individual plastics. The energy footprint of the dissolution-precipitation process of various plastic types was estimated based on carbon content (Smeaton, 2021; Vollmer et al., 2020). The efficiency of the chemical additive extraction technique is varied between 0% and 90%, where 0% signifies a complete bypass of processing the recycled materials. The domestic plastic recycling and international export rates were held constant at 3.9% and 4.5%, respectively. The mechanical recycling efficiency was held constant at 66.7%. Incineration and landfilling remained at a 17.2:82.8 ratio. A plastic loss rate of 10% was assumed post-dissolution-precipitation process.

Fig. 6c demonstrates that increasing the chemical additive extraction rate could effectively prevent over 350,000 tons (317.5 million kg) (95%) of the chemical additives in the recycled plastics from contaminating materials in the subsequent life cycle. The GHG and energy footprint scale steadily with the number of chemical additives removed because of the additional management of the extracted materials as waste. The mixture of extracted chemical additives cannot be separated into individual components economically at a large scale. However, the extraction solvent in the dissolution-precipitation step can be recovered and reused (Aboagye et al., 2021; Chea et al., 2020). The amount of additives released from plastics has been defined as the total plastic additives removed, which scales with the amount of mass of plastics processed. The abrupt jump in GHG releases and energy footprint beyond the 0% chemical additive extraction rate is attributed to subjecting the recovered plastics to dissolution-precipitation, an additional process with scalable GHG emission and energy footprint. Both categories were observed to increase, a necessary trade-off because potentially harmful chemical additives are removed from the recovered plastics. The quality of these materials would approach 95% of the virgin polymer in the best-case scenario. Fig. C3 in the SI summarizes scenario 3. It illustrates the overall movement of plastics, chemical additives, and other MSW at the EoL stages resulting from implementing chemical additive extraction in the post-mechanical recycling stage.

Based on the maximum technical feasibility, these sensitivity analyses have highlighted the merits of improving the existing plastic recycling rate if chemical recycling and additive extraction are implemented before plastic reuse. Incineration is effective at thermally degrading plastics and additives, but these materials are transformed into a non-usable state. Alternatively, landfilling contains plastics and additives with a barrier. However, there are chances of additives being released from the containment in leachates. Although this release is not as high as mechanical recycling, landfilling is not a sustainable solution because these materials take decades and centuries to degrade.

3.8. Potential improvement

Modifications to the current plastic life cycle can be made to minimize resource consumption, environmental impacts, and adverse health effects on humans and animals. The manufacturing, use, collection, and sorting stages would remain consistent with the existing plastic processing methods. However, incentives and policies can be implemented to minimize chemical release. Manufacturers should design plastics with consideration for the possible release of toxic chemical additives during use and EoL stages. For instance, single-use plastics may not necessarily require the same concentration of chemical additives as plastics that play a longer-term role in our daily lives, such as electronics, storage, and automobile parts. Manufacturers may choose to be accountable for the materials produced by designing the materials to be degradable in the presence of a specific substance. Consumers should be informed of the potential danger of subjecting plastics to conditions that favor chemical additive migration during the use stage. In addition, only 8.4% of the plastics collected in the EoL stage were reported as recycled (4.5% exported overseas and 3.9% recycled domestically). Although EoL plastic export is beyond the scope of our study, modifications to domestic and international plastic waste management remain crucial.

Heavy reliance on incineration and landfilling presents the largest obstacle to achieving a circular economy in the plastic life cycle. A circular economy can be achieved if plastic manufacturers design plastics with recycling in mind. Prioritizing plastic recycling practice at the end-of-life stages by reducing incineration, landfilling, and littering may keep plastics in the loop. However, such a condition is unrealistic until technological efficiencies, operation costs, incentives, and legislative support associated with the current plastic recycling efforts are improved (Barra and Leonard, 2018; Hahladakis et al., 2018; Paletta et al., 2019). Smith et al. (2022) performed a separate material flow analysis, specifically on PET throughout the end-of-life processes and further highlighted processing challenges regarding contaminants and additives present during mechanical recycling. One solution to addressing this issue is implementing a chemical additive removal stage to eliminate the primary contamination source. This method may use solvent extraction and dissolution-precipitation to separate the polymer from the chemical additives loosely held in the polymer matrix. The solvent utilized in this process can be recycled and reused (Walker et al., 2020). In addition to mechanical recycling, chemical recycling, such as pyrolysis, may depolymerize the plastics and recover the original monomer and other pyrolysis-derived products such as aromatics, fuels, and waxes (Gracida-Alvarez et al., 2019). This processing route provides opportunities for upcycling, transforming the material into new products with minimal waste (La Rosa et al., 2018; Li et al., 2015, 2022; Meys et al., 2020; Ügdüler et al., 2020). The judicious use of chemical additives during the plastic manufacturing stage and the standardization of material collection, sorting, and recycling (mechanical, chemical, and chemical additive extraction) in EoL plastic processing is expected to address current concerns.

4. Conclusions

Post-consumer plastics are prone to negatively impacting the environment through many forms of unintentional releases. Although highly beneficial for enhancing the properties of the materials, the chemical additives are not linked to the plastics network, allowing the molecules to migrate and contaminate the surrounding environment. The material flow analysis performed in this work has shown that the current plastic economy is linear because most of the plastic waste generated has been incinerated and landfilled. Non-recyclable plastics are sent to incinerators and landfills in large quantities due to sorting and reprocessing challenges. Furthermore, the generic scenario analysis highlighted that incineration and landfilling could adversely affect the environment due to the potential release and exposure of toxic substances to workers and the surrounding areas. The current mechanical recycling methods are not immune to these releases either, which created the need for an improvement in the plastic waste management infrastructure. An environmental impact assessment was completed to predict the environmental consequences of the existing EoL practices by assessing energy footprints, greenhouse gas (GHG) releases, and ecological footprints. This result can help identify the environmental burden and motivate the shift toward a circular economy approach with minimal releases. We provided a theoretical representation of the area required to achieve one unit of product (recycled plastics) sustainably. The plastic waste processing in MSW requires up to 7% of the total arable land area on Earth. This theoretical area should be minimized because land is a limited resource, and it is not sustainable to consume resources faster than the generation rate. In this case, the treatment of plastic waste and mechanical recycling are hot spots that cost the greatest theoretical ecological footprint. Thus, we further emphasized the need to optimize the plastic waste treatment process and minimize the chemical additive releases and contamination that were identified from the material flow analysis and generic scenario analysis.

We have also demonstrated from our sensitivity analyses that mechanical recycling may assist with closing the material loop, despite the large ecological footprints. Poor recycling efficiency caused by chemical additive contamination and migration can further create the opportunity for toxic substances to release into the surrounding environment. However, the recycling rate can realistically be improved, reducing the reliance on incineration and landfilling, if initiatives are taken to enhance EoL plastic management. More importantly, plastic recycling should not be the sole focus. Chemical recycling should be used to process waste plastics that cannot be recycled mechanically. Additionally, chemical additive extraction should be used with conventional mechanical recycling to minimize unintentional chemical releases and contamination in the future life cycle. Ultimately, the effective separation of plastics and additives without compromising the structural and mechanical properties of the original product is the key to making plastic recycling safer and more sustainable.

Therefore, analysis regarding chemical additive releases during chemical recycling should be investigated to determine a mitigation strategy before a standard chemical recycling practice is used. We aim to create a tool that performs all the calculations done in this work while providing the users with crucial data and plots. This tool can simulate the effects of altering key parameters (recycling, incineration, landfilling rate, MSW composition, import and export rate, and efficiency at various stages) on chemical releases, greenhouse gas emissions, and energy footprint. Although the data used in this work is primarily from 2018, this tool is also set to contain data from previous years, allowing users to generate a custom dataset to test their own scenarios. This contribution may be used to support the United Nations Sustainable Development Goals for ensuring sustainable consumption and production patterns with plastic waste.

Environmental implication

The U.S. generated over 35.7 million tons of plastic waste in 2018. These plastics contain 18,000 – 25 million tons of chemical additives, from which around 200,000 – 1 million tons exit the plastic end-of-life (EoL) stage into the environment. EoL activities like recycling, energy recovery, and landfilling release hazardous chemical additives that negatively affect the environment and human health. This article describes a novel analysis of the U.S. EoL stage of plastics additives. We perform a qualitative and quantitative tracking of chemical additives to estimate their potential environmental releases, life cycle inventories, occupational exposure, and environmental impacts within the plastic EoL.

HIGHLIGHTS.

• Material flow analysis estimates chemical additive releases throughout plastic end-of-life (EoL).

• Generic scenario analysis identifies and quantifies releases and occupational hazards in plastic EoL stage.

• Maximizing recycling can reduce uncontrolled chemical additive releases.

• Combining chemical recycling with mechanical recycling increases upcycled plastic quality.

Data Availability

I have shared the link to my data in the manuscript.