Issuu

Report of the PlastLIFE project coordinated by the Finnish Environment Institute

Methodological considerations in plastics-related life cycle assessment Tomas Ekvall

Report of the PlastLIFE project coordinated by the Finnish Environment Institute

Methodological considerations in plastics-related life cycle assessment Tomas Ekvall

Report of the PlastLIFE project coordinated by the Finnish Environment Institute Methodological considerations in plastics-related life cycle assessment Author: Tomas Ekvall, TERRA Co-funder: EU LIFE SIP - funding program Publisher: Finnish Environment Institute (Syke), syke.fi/en Cover: stock.adobe.com The publication (pdf) is available in the internet: plastlife.fi (In English) > Circular economy research > Publications about circular economy and Finnish Environment Institute publications archive (helda.helsinki.fi) > Syke-hankkeiden julkaisuja ISBN 978-952-11-5866-7 (pdf) Year of issue: 2026

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

Abstract Methodological considerations in plastics-related life cycle assessment This report includes guidance for dealing with key issues in the assessment of environmental impacts of products and processes within the framework of the PlastLIFE project, which aims to establish a safe and sustainable circular economy for plastics in Finland by 2035. The starting point for the assessment methodology is the Product Environmental Footprint (PEF) framework developed by the European Union. The report clarifies this framework and discusses whether deviations from the PEF methodology should be made in the PlastLIFE project. The focus is on key issues identified in a dialogue with project participants, such as the assessment of littering and microplastics, the choice between attributional and consequential LCA, co-product allocation, waste management and recycling, dynamic modelling, electricity, and chain-of-custody (CoC) approaches. The report discusses the pros and cons of different methodological options but leaves the final decision on methods to the LCA practitioner in PlastLIFE. The PEF framework should be considered the default approach and deviations from this framework should be transparently reported and justified. Further research is warranted on most of these methodological issues. It would be particularly interesting to: • investigate when the Circular Footprint Formula (CFF) can be replaced by a simpler method without loss of important information (see Section 4.2); • calculate Factor B in the CFF based on the equation B = VE/(VE+VGF) (Section 5.4); • use the premise database tool for modelling of future processes (Section 6.1); • refine and test a discounting approach for modelling delayed GHG emissions (Section 6.2); • analyze the similarities and differences between CoC models for electricity and materials (Chapters 7-8); and • analyze how diverging views on the role of LCA affects methodological choices (Chapter 9). Keywords: life cycle assessment, environmental footprints, plastics, sustainability, circularity, methodology

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

Tiivistelmä Metodologisia näkökohtia muoveihin liittyvässä elinkaariarvioinnissa Tämä raportti sisältää ohjeita tuotteiden ja prosessien ympäristövaikutusten arviointiin. Raportti on laadittu PlastLIFE-hankkeen puitteissa, jonka tavoitteena on luoda turvallinen ja kestävä kiertotalous muoville Suomessa vuoteen 2035 mennessä. Ympäristövaikutusten arviointimenetelmän lähtökohtana on Euroopan unionin ympäristöjalanjälkilaskennan (Product Environmental Footprint, PEF) viitekehys. Raportissa tarkennetaan tätä viitekehystä siten, että se sopii PlastLIFE-hankkeen ja muovien kontekstiin. Raportin painopiste on hankkeen osallistujien kanssa käydyssä vuoropuhelussa tunnistetuissa keskeisissä elinkaariarvioinnin menetelmällisissä kysymyksissä, kuten roskaantumisessa ja mikromuovien arvioinnissa, attribuutioperusteisen ja seurauksellisen elinkaariarvioinnin mallinnuksen valinnassa, sivutuotteiden kohdentamisessa, jätehuollossa ja kierrätyksessä, dynaamisessa mallinnuksessa, sähkön tuotannon päästöjen arvioinnissa sekä alkuperäketjun (chain-of-custody) mallin määrittelyssä. Raportissa käsitellään eri menetelmävaihtoehtojen hyviä ja huonoja puolia. Lopullinen päätös menetelmistä jätetään PlastLIFE-menetelmän elinkaarianalyysin suorittajan tehtäväksi. PEF-kehystä tulisi pitää ensisijaisena lähestymistapana, ja poikkeamat tästä kehyksestä tulisi raportoida ja perustella avoimesti. Useimpien näiden metodologisten kysymysten osalta tarvitaan lisätutkimuksia. Erityisen mielenkiintoisia lisätutkimuksen aiheita olisivat: • tutkia, milloin the Circular Footprint Formula (CFF) voidaan korvata yksinkertaisemmalla menetelmällä menettämättä tärkeää tietoa (katso luku 4.2); • laskea CFF:n tekijä B yhtälön B = VE/(VE+VGF) perusteella (luku 5.4); • käyttää lähtötietokantatyökalua tulevien prosessien mallintamiseen (luku 6.1); • tarkentaa ja testata diskonttausmenetelmää viivästyneiden kasvihuonekaasupäästöjen mallintamiseen (luku 6.2); • analysoida sähkön ja materiaalien alkuperäistarjousmallien yhtäläisyyksiä ja eroja (luvut 7–8); ja • analysoida, miten elinkaarianalyysin roolia koskevat erilaiset näkemykset vaikuttavat menetelmävalintoihin (luku 9). Asiasanat: elinkaariarviointi, ympäristöjalanjälki, muovit, kestävä kehitys, kiertotalous, metodologia

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

Sammandrag Metodmässiga överväganden vid plastrelaterad livscykelanalys Denna rapport innehåller vägledning för viktiga metodval vid bedömning av miljöpåverkan från produkter och processer inom ramen för PlastLIFE-projektet, vars syfte är att etablera en säker och hållbar cirkulär ekonomi för plast i Finland senast 2035. Utgångspunkten för bedömningsmetodiken är EUs ramverk för livscykelanalys (LCA): Product Environmental Footprint (PEF). Rapporten förtydligar detta ramverk och diskuterar huruvida avvikelser från PEF-metodiken bör göras i PlastLIFE-projektet. Fokus ligger på viktiga frågor som identifierats i en dialog med projektdeltagarna, såsom bedömning av nedskräpning och mikroplaster, valet mellan bokförings- och konsekvens-LCA, allokering vid samproduktion, avfallshantering och återvinning, dynamisk modellering, elenergi, samt spårbarhetsmetoder. Rapporten diskuterar för- och nackdelar med olika alternativa metoder men lämnar det slutgiltiga metodvalet till de som utför livscykelberäkningarna i PlastLIFE. PEF-metodiken bör dock användas som default och avvikelser från detta ramverk bör rapporteras och motiveras transparent. Ytterligare forskning är motiverad kring de flesta av dessa metodfrågor. Det vore särskilt intressant att: • undersöka när Circular Footprint Formula (CFF) kan ersättas med en enklare metod utan förlust av viktig information (se avsnitt 4.2); • beräkna faktor B i CFF baserat på ekvationen B = VE/(VE+VGF) (avsnitt 5.4); • använda databasen premise för modellering av framtida processer (avsnitt 6.1); • förfina och testa en diskonteringsmetod för modellering av fördröjda växthusgasutsläpp (avsnitt 6.2); • analysera likheter och skillnader mellan spårbarhets-modeller för el och material (kapitel 7-8); och • analysera hur olika synpunkter på livscykelanalysens roll påverkar metodval (kapitel 9). Nyckelord: livscykelanalys, miljöavtryck, plast, hållbarhet, cirkularitet, metodik

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

Preface

PlastLIFE is an extensive cooperation project initiated to promote the circular economy of plastics, in accordance with the Plastics Roadmap for Finland. The goal is a safe and sustainable circular economy for plastics in Finland by 2035. This can be achieved through a combination of the following overlapping strategies: • Refraining from unnecessary consumption of plastics • Replacing fossil plastics with bio-based materials and/or other solutions • Reducing littering and other negative impacts of plastics • Increasing recycling of all types of plastics Several technological solutions are developed in the PlastLIFE project, particularly in Work Packages (WPs) 2, 4, 5, and 6, with an aim to contribute to reducing the negative impacts of plastics. Environmental assessments are made to investigate if they reduce climate and environmental impacts. The results of the assessment will depend on the methods applied. Hence, there is a need for a methodological framework for the environmental assessments. This document is produced within Task 9.4.1 of the project. It presents discussions and recommendations on a range of key issues in the methodology for assessments of products and processes. This is also a step towards the goal to create a national framework for assessing the circular economy of plastics in Finland, which is part of Task 9.4.3. The methodological considerations in this report were developed in cooperation with LCA experts at the Finnish Environment Institute (Syke), and with researchers from LUT University and Natural Resources Institute Finland (LUKE). However, the author of this report is solely responsible for the conclusions and recommendations presented.

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

Contents

Methodological considerations in plastics-related life cycle assessment Abstract .................................................................................................................................................. 3 Tiivistelmä .............................................................................................................................................. 4 Sammandrag .......................................................................................................................................... 5 Preface ................................................................................................................................................... 6 Contents ................................................................................................................................................. 7 1

Introduction..................................................................................................................................... 9 1.1 Basic principles for the methodology ....................................................................................... 9 1.2

Related work ........................................................................................................................... 11

Littering and microplastics ....................................................................................................... 12

Attributional or consequential modelling ............................................................................ 14

Allocation rules ............................................................................................................................ 16 4.1 Allocation at multifunctional processes ................................................................................. 16 4.2

Allocation at waste management ........................................................................................... 17

Substitution in the CFF ................................................................................................................ 19 5.1 The avoided material production (EV*)................................................................................... 19 5.2

Factor A ................................................................................................................................... 19

5.3

The quality ratio (QS/QP) ......................................................................................................... 19

5.4

Factor B ................................................................................................................................... 21

Accounting for time.................................................................................................................... 24 6.1 Prospective LCA ...................................................................................................................... 24 6.2

Temporal storage and release ................................................................................................ 25

Electricity modelling ................................................................................................................... 29

Mass-balance and book-and-claim .................................................................................... 31

Outlook .......................................................................................................................................... 34

References ............................................................................................................................................ 36

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

1 Introduction

1.1 Basic principles for the methodology The PlastLIFE project develops solutions aiming for a safe and sustainable circular economy for plastics in Finland. These can involve refraining from unnecessary consumption of plastics, replacing fossil plastics with bio-based materials and/or other solutions, reducing littering and other negative impacts of plastics, and/or increasing recycling of all types of plastics An environmental assessment of such circular solutions for plastics needs a broad systems perspective to account for potentially important system impacts. It is, for example, vital to account not only for the impacts of recycling and repair processes, but also for the avoided production of substituted primary materials and products and for the avoided waste incineration and landfills. Life cycle assessment (LCA) and carbon footprint calculations are tools for environmental assessment with a wide systems perspective. An LCA is a compilation and evaluation of the inputs, outputs and potential environmental impacts of a product from cradle to grave, i.e. from raw material acquisition or generation from natural resources, through production, distribution and use of the product, to final disposal. Carbon footprints share the same life cycle perspective but are limited to the climate impacts of greenhouse-gas emissions throughout the life cycle of the product. The concept “product” is in this context interpreted broadly: it includes also materials and services. Life-cycle calculations and life-cycle thinking are well established in both industry and policy-making. For the past decades, they have been integrated into many pieces of EU legislation on waste management, resource efficiency, and elsewhere. The European Commission also developed a framework for LCA: Product Environmental Footprint (PEF; EC 2021). This is the starting point for our methodological considerations. The PEF initiative is an attempt to develop and present an LCA methodology fit for many purposes. This is a difficult challenge, since each purpose or application of LCA comes with specific requirements on the method. Draft EU legislation requires that producers or importers of batteries or photovoltaic equipment apply the PEF methodology to demonstrate that the carbon footprint of their products does not exceed stipulated thresholds. Results from PEF calculations can also be used for marketing purposes. These applications require that the methodology is robust and not susceptible to subjective methodological choices or assumptions. Aspects that are relevant but uncertain is best excluded from such calculations. Our aim, in contrast, is to assess technologies and systems to give a basis for policy decisions. In this application, the calculation should be comprehensive: it should account relevant aspects even if the information is uncertain. When the PEF methodology is used to assess many products, a streamlined approach is valuable. This can include, for example, default data on the production of various materials, etc. The drawback of such data is that they do not allow for assessing improvements in the production processes. Such assessment requires the use of process- or site-specific data rather than default data. Hence, to be used for assessments in PlastLIFE, the PEF framework needs to be adapted to the PlastLIFE context. This can require clarifications when the framework is open for different

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

interpretations and/or allows for a free choice of methods. Guidance for the application of the framework is also likely to be needed on some points. In addition, the PlastLIFE methodology might deviate from PEF where this is called for to assess the proposed solutions and guide decisions towards an environmentally sustainable future. Deviations will be justified, for example, when the PEF framework produces results that guide decisions in the wrong direction, when it is too cumbersome to be feasible, or when it is not specific enough to allow for assessing specific solutions. However, our aim is not to present a standardized method to fit in all PlastLIFE case studies. What methods are applicable in the environmental assessments not only varies between cases but also depends on the subjective perspective of the researcher. The aim of this work is not to restrict the academic freedom of the PlastLIFE researchers, but instead to give support in their methodological choices.

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

1.2 Related work This report complements other efforts to develop methods based on PEF, such as: • The EU Joint Research Center (JRC) has published detailed rules, based on PEF, on how to conduct LCA studies of plastic products from different feedstocks: the Plastics LCA Method (Nessi et al. 2021). • An attempt has been made to develop PEF Category Rules (PEFCR) for flexible packaging. These should include detailed rules on how to apply PEF to such packaging. The attempt to develop a formal PEFCR was abandoned after the first public consultation. However, the actors involved still aim to develop a “shadow PEFCR”. • The EU project ORIENTING (www.orienting.eu) ran from November 2020 to April 2024 with the aim to develop a robust and operational method for life cycle sustainability assessment (LCSA) of products and services (Horn et al. 2021). The environmental assessment methods builds on PEF but expands on the land-use assessment, where PEF applies the Land Use Indicator Value Calculation in Life Cycle Assessment (LANCA) that focuses on a Soil Quality Indicator; Orienting in addition accounts for impacts on biodiversity (through the Biomap approach) and bioresources. Their LCSA framework also includes indicators for circularity and criticality. • The EU project CALIMERO (calimeroproject.eu) started summer 2022 with the goal to create a common framework for the LCSA of biobased products. They use PEF as a starting point for the environmental assessment but aim to include also approaches to assess temporal carbon storage, specific characterization factors for toxicity, impacts on biodiversity and ecosystems services, and indicators for circularity and criticality. • The EU project NOVAFERT (www.novafert.eu) started in September 2022 with the goal to guide production and use of alternative fertilizing products according to the best environmental performance. For this purpose, they aim to develop an assessment method based on PEF. The EU project ALIGNED (https://alignedproject.eu) started in October 2022 and aims to harmonize and advance the scientific use of LCA in the bio-based sector. This project differs from the others in that it does not use PEF as a starting point. Instead, it aims to refine and expand on consequential LCA. It focusses on, for example, how to model competition over land and biomass, carbon accounting (incl. modelling of temporal carbon storage), indirect land-use change, and impacts on biodiversity. They also integrate methods for prospective LCA, i.e., LCA of future products and systems. In addition, the Finnish PLASTin project ran from 2020 to 2022 and aimed to support the plastic industry actors to develop systemic, and environmentally optimized recycling concepts. This was achieved with the new knowledge on recycling processes and on technologies for, e.g., sorting, pretreatment, mechanical and chemical treatment and reject handling, and about system-level understanding, allowing improved business opportunities based on recycling.

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

2 Littering and microplastics

Litter has an aesthetic impact on our environment. It also poses a hazard for animals that can get entangled in the litter. When the litter is plastic these impacts remain for a long time, because plastics degrade very slowly. When plastic litter degrades, it forms secondary microplastics (with sizes from 1 μm to 5 mm) and/or nanoplastics (1 nm to 1 μm; Frias & Nash 2019). Part of these particles find their way to lakes and to the sea together with primary microplastics, which are directly released to the environment as small particles. Marine animals can mistake such particles for food, thus reducing their intake of nutritious food. Through food chains the particles can find their way to humans. Additives in plastic particles, such as stabilizers and flame-retardants, may be harmful to animals or humans ingesting them (EU 2025). The Economist (2025) reports that the many possible impacts of microplastics are difficult to study and, hence, uncertain. However, Landrigan et al. (2025) argue that the dangers of plastics are grave and under-recognised. The problems caused by microplastics are specific to environmental assessments of plastic products. Littering and microplastics are not important environmental aspects for plastic products that do not end up as litter or primary microplastics in nature. Even when they do, this might not be an important environmental problem in all cases. Salieri et al. (2021) found, using worst-case assumptions, that microplastics from a polyester t-shirt and a shower gel had only a small impact on freshwater toxicity. However, littering and microplastics are the main environmental drawbacks of plastic carrier bags, compared to other carrying solutions. An investigation made within PlastLIFE indicates that the most common items of plastic litter near underground stations in Helsinki are cigarette butts (which contain the plastic cellulose acetate), used snuff bags (which can contain nylon and other plastics), and plastic packaging. Laundering of synthetic clothes and the use of car tires are important sources of primary microplastics (EU 2025), together with granules from artificial turf (Brodén 2022). Secondary microplastics originate from, for example, plastic bags, bottles, and fishing nets (EU 2025). Hence, microplastics can be important to account for in LCAs of at least these products. Early attempts to quantify the impacts of littering in LCA were made by, e.g., ExcelPlas Australia (2004) in a study on plastic bags, assuming that 0.5 % of the investigated bags end up as litter (ibid. p.70). The aesthetic impact of the litter was quantified by multiplying the surface area of the litter by the assumed time it takes to degrade. The indicator for impacts of litter on marine fauna (through entanglement or ingestion) was calculated by multiplying the weight of the litter by the time it stays afloat. The latter was assumed to be 6 months for traditional plastics and 3 months for degradable plastics. Edwards & Parker (2012) used the same approach for quantifying aesthetic impacts of litter in an LCA of carrier bags in the UK; however, they assumed 0.75% of the bags ends up as litter (ibid. p.12). Civancik-Uslu et al. (2019) quantified the littering potential of different carrier bags by accounting for the weight, degradability and price of the bags. They found that the price is an important indicator that influences the probability of the bags becoming litter. The database ecoinvent includes a rough estimate of the quantity of litter by based on a “littering event probability” for each flow and process. This probability can vary from 0% to 95% (Ciroth & Kouame 2019).

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

The Plastic Leak Project developed an approach to estimate the quantity of plastics ending up as litter (macroplastics) in oceans or in terrestrial environment, or as microplastics in oceans, freshwater sediment, or terrestrial environment (Peano et al. 2020). This approach is included as an option in the Plastics LCA Method, i.e., the PEF methodology that Nessi et al. (2021) recommends for assessing plastics. Maga et al. (2022) presented a method for calculating the residence time of various plastics objects in different environmental compartments. The MarILCA working group takes the assessment further by estimating impacts; however, it is limited to two impacts in the marine environment: entanglement caused by macroplastics and physical impacts of microplastics (Lavoie et al. 2021; Woods et al. 2021; Corella-Puertas et al. 2022; Høiberg et al. 2022; Loubet et al. 2022; Corella-Puertas et al. 2023). An alternative or complementary method might be to include in the life cycle inventory analysis (LCI) the energy use and emissions related to cleaning the ground and water from litter. At a workshop for PlastLIFE Task 9.4 in Helsinki in November 2023, we found the method of Plastic Leak Project interesting, because it is part of the PEF framework through the Plastics LCA Method. If possible, we should also model the potential impacts of the litter. It might be difficult, however, to find input data for such a model.

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

3 Attributional or consequential modelling

The distinction between attributional and consequential LCA arose to resolve debates on what input data to use in an LCA and how to deal with allocation problems. An attributional life cycle assessment (ALCA) is based on average data and allocation is performed by partitioning environmental burdens of a process between the life cycles served by this process (Tillman 2000). A consequential LCA (CLCA), in contrast, ideally use marginal data in many parts of the life cycle and avoids allocation through system expansion. The benefit of keeping these two types of LCA distinct is that they respond to different questions (see Figure 1). An ALCA estimates what share of the global environmental burdens that belongs to a product. A CLCA gives an estimate of how the global environmental burdens are affected by the production and use of the product.

Fig 1. Illustration of how attributional and consequential LCA relate to the total environmental burdens in the world – represented by a circle – and in what sense they respond to different questions. Based on Weidema (2003). The PEF methodology includes a mix of attributional and consequential elements. For example, it takes a consequential approach to the modelling of waste management and recycling (Section 5), but an attributional approach to electricity modelling (Section 7). This means that PEF results do not respond to a clear question, other than “What is the Environmental Footprint of this product”? This might be

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

good enough, if we stick to the PEF methodology; however, if we introduce deviations from the PEF methodology, our assessment results will not even respond to that question. Hence, if we decide to deviate from the PEF methodology, we should consider whether to make the approach more consistently attributional, or more consistently consequential. Alternatively, we might find other ways to describe the methodology and the meaning of the results. The clarity of the concepts is reduced by the fact that different LCA experts disagree on what attributional and consequential LCA means. The text above is based on a specific interpretation of these terms. We should decide on an interpretation for the project and clearly present this interpretation as well as our justifications for it. A starting point could be the suggestion recently presented by Ekvall (2023). We discussed this issue at the November workshop in Helsinki. We found that the combination of average data and substitution applied in PEF can be adequate; however, a more consistently consequential modelling of waste management has been established in part of the group. Based on the above, I recommend to allow for a diversity of methods, making it possible for PlastLIFE researchers to choose the approach they prefer.

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

4 Allocation rules

4.1 Allocation at multifunctional processes The Plastics LCA Method (Nessi et al. 2021) and the PEF methodology in general, includes a hierarchy of preferred approaches to allocation problems: 1. Subdivision, or system expansion 2. Allocation based on underlying physical relationships, or direct substitution 3. Allocation based on other relationships, or indirect substitution The international standard for LCA (ISO 14044) includes a similar hierarchy for allocation. However, the hierarchies differ on one important point: the interpretation of system expansion as a way to avoid allocation, which both documents state should be done when possible. A descriptive annex on allocation has been added in the most recent version of ISO 14044 (ISO 2020a), where system expansion is stated to involve substitution, which means that the expanded system will have a single net functional output. The PEF guidance, in contrast, interprets system expansion to result in a system with multiple functional outputs (see Figure 2). In a waste-management system this means that the options investigated and compared treat the same quantity of waste, produce the same amount of material (through recycling or primary production) and generate the same amount of energy (through, e.g., waste incineration or from primary energy sources.

Fig 2. Two versions of system expansion at a multifunctional process. Based on Tillman et al. (1994). ISO (2020a) states that the two interpretations of system expansion are equivalent, but PEF clearly distinguishes between them. System expansion to multiple functions is presented as a way to avoid allocation, but system expansion with substitution is presented as a way to perform allocation. The latter should be done only when the former is not feasible or applicable.

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

The PEF hierarchy of approaches to allocation presents system expansion with substitution on par with the conventional methods for partitioning environmental burdens and impacts. This means the PEF practitioner is free to choose between substitution and partitioning. Based on the above, I recommend keeping this flexibility of approaches to allocation. Partitioning fits well in attributional assessments. Substitution is often an appropriate choice in consequential assessments.

4.2 Allocation at waste management The rules for modelling recycling and waste management are more specific; they are given by the Circular Footprint Formula (CFF), which can be divided into three parts: Material: (𝟏𝟏 −𝐑𝐑𝟏𝟏)×𝐄𝐄𝐕𝐕 +𝐑𝐑𝟏𝟏×(𝐀𝐀×𝐄𝐄𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫 +(𝟏𝟏−𝐀𝐀)×𝐄𝐄𝐕𝐕×𝐐𝐐𝐒𝐒𝐒𝐒𝐒𝐒/𝐐𝐐𝐩𝐩)+(𝟏𝟏−𝐀𝐀)×𝐑𝐑𝟐𝟐×(𝐄𝐄𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫𝐫− 𝐄𝐄𝐕𝐕∗×𝐐𝐐𝐒𝐒𝐒𝐒𝐒𝐒𝐒𝐒/𝐐𝐐𝐏𝐏) Energy:

(𝟏𝟏 −𝐁𝐁)×𝐑𝐑𝟑𝟑×(𝐄𝐄𝐄𝐄𝐄𝐄−𝐋𝐋𝐋𝐋𝐋𝐋×𝐗𝐗𝐄𝐄𝐄𝐄,𝐡𝐡𝐡𝐡𝐡𝐡𝐡𝐡×𝐄𝐄𝐒𝐒𝐒𝐒,𝐡𝐡𝐡𝐡𝐡𝐡𝐡𝐡−𝐋𝐋𝐋𝐋𝐋𝐋×𝐗𝐗𝐄𝐄𝐄𝐄,𝐞𝐞𝐞𝐞𝐞𝐞𝐞𝐞×𝐄𝐄𝐒𝐒𝐒𝐒,𝐞𝐞𝐞𝐞𝐞𝐞𝐞𝐞) Disposal:

(𝟏𝟏−𝐑𝐑𝟐𝟐−𝐑𝐑𝟑𝟑)×𝐄𝐄𝐃𝐃

The parameters in the CFF have the following meaning: • A: allocation factor of burdens and credits between supplier and user of recycled materials. • B: allocation factor of energy recovery processes. It applies both to burdens and credits. • QSin: quality of the ingoing secondary material, i.e. the quality of the recycled material at the point of substitution. • QSout: quality of the outgoing secondary material, i.e. the quality of the recyclable material at the point of substitution. • QP: quality of the primary material, i.e. quality of the virgin material. • R1: the proportion of material in the input to the production that has been recycled from a previous system. • R2: the proportion of the material in the product that will be recycled (or reused) in a subsequent system. Therefore, R2 shall take into account the inefficiencies in the collection and recycling (or reuse) processes. R2 shall be measured at the output of the recycling plant. • R3: the proportion of the material in the product that is used for energy recovery at end of life (EoL). • Erecycled: specific emissions and resources consumed (per functional unit) arising from the recycling process of the recycled (reused) material, including collection, sorting and transportation process. • ErecyclingEoL: specific emissions and resources consumed (per functional unit) arising from the recycling process at EoL, including the collection, sorting and transportation processes. • EV: specific emissions and resources consumed (per functional unit) arising from the acquisition and pre-processing of virgin material. • EV*: specific emissions and resources consumed (per functional unit) arising from the acquisition and pre-processing of virgin material assumed to be substituted by recyclable materials.

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

• • • • •

EER: specific emissions and resources consumed (per functional unit) arising from the energy recovery process (e.g. incineration with energy recovery, landfill with energy recovery, etc.). ESE,heat and ESE,elec: specific emissions and resources consumed (per functional unit) that would have arisen from the specific substituted energy source, heat and electricity respectively. ED: specific emissions and resources consumed (per functional unit) arising from the disposal of waste material at the analysed product’s EoL, without energy recovery. XER,heat and XER,elec: the efficiency of the energy recovery process for both heat and electricity. LHV: lower heating value of the material in the product used for energy recovery.

This formula stipulates that substitution should be done both at recycling of material (where the impacts of the substituted material production is given by EV*) and at recovery of energy (ESE,heat and ESE,elec). The guidelines for Product and Organisational Environmental Footprints stipulate that the CFF should be used to model end-of-life in any LCA, and also for modelling the input of recycled material and energy from waste. However, the CFF can be difficult to apply. In assessments of complex products and processes that involve many different materials, the formula is cumbersome. In many cases, the waste management or recycled content of a specific material also has little impact on the assessment results. This means that simplifications to the formular can be not only justified but also important to make the assessment feasible. There are also cases where the CFF is not sufficiently comprehensive. It neglects impacts that can potentially be significant for the PEF results and conclusions. One example is that the use of recycled plastics in a production process can reduce the need for waste-plastics disposal and its associated environmental impacts. Another example is that recycling of plastics to substitute wood or concrete affects not only the substituted material production, but also the waste management at the end-of the product where the recycled plastics is used; the climate impact of incinerating a product produced from recycled plastics are much greater than the climate impacts of incinerating the competing wood product (Ekvall et al. 2024). In addition, there are several technical details in the CFF that should be discussed (see next section). Based on this, my recommendation is to use the CFF as the default method, but to allow for deviations when this is justified. A simpler approach, such as the cut-off, is clearly justified when this does not significantly affect the assessment results and does not affect the conclusions. It can also be justified if and when the CFF is too difficult to apply. On the other hand, the CFF should be expanded to account for, e.g., consequences for the waste management of upstream or downstream products in the recycling cascade, when these consequences are important for the conclusions of the assessment.

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

5 Substitution in the CFF

5.1 The avoided material production (EV*) The PEF rules and the Plastics LCA Method stipulate that the credit given at recycling of material from the product life cycle should be a share (specified by the allocation Factor A and the quality ratio QS/QP) of the emissions and resources consumed at the acquisition and pre-processing of the virgin material assumed to be substituted (EV*; EC 2021, p. 251-252; Nessi et al. 2021, p. 91). A recycled material might not substitute a similar material, particularly if the quality loss in the recycling is large. Downcycled plastics might, for example, substitute wood or concrete. Since wood has a completely different environmental profile compared to plastics, it might not be sufficient to adjust the quality factor QS/QP to account for the downgrading of the recycled material. The calculation should instead include the production of the material that is actually substituted. This is also what the PEF methodology stipulates: when evidence can be provided to support the assumption that the substituted material production (EV*) is different from the original primary-material production (EV), then EV* should represent the production of the type and amount of the actually substituted material (EC 2021, p. 258; Nessi et al. 2021, pp. 83-84). My recommendation is to adhere to the PEF rules in this matter: EV* should represent the production of the actually substituted material, when there is evidence to support this.

5.2 Factor A The substitution means that part of the environmental net benefit of recycling is assigned to products that are recycled after use. This part is quantified through a material-specific allocation Factor A and adjusted for losses in quality (Q). Factor A can in the PEF framework take the values 0.2, 0.5, and 0.8. A low value on Factor A (0.2) means that the collection and supply of recyclable material is the most important for increasing total recycling of the material. A high value (A=0.8) means that the demand for and use of recycled material, i.e., the recycled content of the product, is the most important. The PEF framework includes default values for common materials: it is 0.2 for polypropylene used in lead-acid batteries, but 0.5 for all other plastics. It is also 0.5 for all materials where a default value is lacking. My recommendation is to use the PEF default values for Factor A, unless clear evidence regarding the supply and/or demand for a recyclable material gives a basis for a different Factor A.

5.3 The quality ratio (QS/QP) The quality ratio between recycled and primary material (QS/QP) is mainly relevant to apply when a recycled material substitutes a similar kind of material, for example when recycled plastics substitutes primary plastics. In fact, the PEF rules (EC 2021, p. 52) state that the quality factor QS/QP should only be

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

applied to material recycled from the life cycle when the environmental impacts of the substituted material (EV*) is the same as the environmental impacts of the primary production of the material in the product (EV). This is a harsh condition, because two production processes rarely have the same environmental impacts even when they produce the same material. The PEF guidance states that the quality ratio (QS/QP) by default is 0.9 for most plastics. Exceptions are made for: • PET recycled in a closed system, where the default quality ratio is 1, and • LDPE film, where the default quality ratio is 0.75. The PEF framework allows for deviations from the default quality ratios when there is evidence to support other values. In general, the quality ratio of recycled to primary materials should be defined by the price; however, when the price of recycled material is higher than the price of primary material the quality ratio is set to 1.0. When physical aspects are more relevant than the price, the quality ratio can be estimated based on physical properties of the material (EC 2021, p. 254). When assessing methods for recycling that differ in the quality of the recycled material, it is necessary to be able to distinguish in the CFF between recycled material of different quality. Hence, the PlastLIFE project might need an approach for quantifying the quality as a complement to the default value QS/QP =0.9. The scientific literature includes several methods for quantifying the quality in recycled plastics and other materials based on physical properties. Vadenbo et al. (2017) present a framework that several other researchers refer to. In the framework, an end-use-specific functionality of a recycled material (φrec) is defined by: φrec = prec,tech · min (frec,tech, frec,inst, frec,user), where prec,tech

is a technical property of the recycled material, and

frec,tech, frec,inst, frec,user are requirements based on technology, regulations, and users This framework starts with a single physical property but can be expanded to account for multiple properties. on a single property in the material. Schulte et al. (2023) propose a framework that accounts for multiple properties by calculating the circularity potential as a weighted average over the relevant properties:

, where wPi

is a weighting factor, and

cPirec

is the extent to which the recycled material meets technical requirement i

Another approach is presented by Roosen et al. (2023). They suggest to quantify the technical suitability for substitution (TSS) in different applications of the recycled material. They then calculate the Virgin Displacement potential of Material j (VDPj) by multiplying the suitability by the market share of each application i (Wmi):

, where

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

EOL-RR is the recycling rate, and EBCs

are economic boundary conditions

We discussed the approaches to quantify quality at the November workshop. We found that a simple approach, using the default value Qs/Qp = 0.9 is often likely to be sufficient. When we need specific values on the quality ratio, we tentatively suggested to still use the price ratio as basis for the quality ratio, with Qs/Qp = 0.9 as the maximum value. This is a slight deviation from the PEF methodology, where Qs/Qp = 1.0 is the maximum value. If relevant, we will account for physical properties and/or a diversity of applications (cf. Roosen et al. 2023) instead of the price ratio. Based on the above, I recommend applying the quality ratio (Qs/Qp) when a recycled material substitutes a similar kind of material, even if the environmental impacts of the substituted material (EV*) is different from the environmental impacts of the primary production of the material in the product (EV). I also recommend to adhere to the findings of the November workshop, with the possible addition that Qs/Qp can be 1 for PET recycled in a closed system.

5.4 Factor B Factor B is the share of the burdens and credits of energy recovery that is assigned to the recovered energy in the CFF. In the current PEF framework, the default value of Factor B is zero. This means that all burdens and credits associated with energy recovery are assigned to the product generating the combustible waste. This corresponds to system expansion with substitution, which is a well-established approach in LCA. Ekvall et al. (2021) point at several problems with this approach: • It is not an easy fit in attributional LCA, because it involves substitution. • It is also inaccurate in consequential LCA, unless the energy recovered from a waste product adds to the energy already recovered. Waste-incinerator plants typically run at maximum capacity, and adding a new waste stream will not increase the energy recovered in the short run. In the long run, a new waste stream can contribute to investments in new incinerator capacity, but this effect is uncertain and can be affected also by the demand for the recovered energy. • It can give an incorrect incentive to incinerate material from a product system when recycling creates the greatest net environmental benefit. This is because the approach assigns the full net benefit of energy recovery to the product system, but only part of the net benefit of recycling. With the default allocation Factor A = 0.5 and the quality ratio Qs/Qp = 0.9, just (1-0.5)*0.9 = 45% of the net benefit of recycling is credited to the recycled product. We need to decide whether this is a case where our methodology should differ from PEF. If so, several alternative approaches can be considered. A purely attributional approach would be to apply economic allocation to the environmental burdens of the energy-recovery process. Economic allocation is based on the revenues from accepting the waste (gate fees; VGF) and energy sales (VE): • VGF/(VE+VGF) is the share of the burdens allocated to the product system where the combustible waste is generated, and • VE/(VE+VGF) is the share of the burdens allocated to the product system where the recovered energy is used. Ekvall et al. (2021) propose two approaches based on consequential thinking to reduce the risk that assessment results point in the wrong direction. One is a short-term consequential approach. It is based on the notion that waste from our product system in the short run displaces other waste streams in the

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

incinerator. The system investigated is expanded to include the alternative treatment of the displaced waste. The second approach proposed by Ekvall et al. (2021) capture part of the long-term consequences. It is based on the notion that investment in new energy-recovery capacity is driven by the expected revenues, which is a reasonable assumption when the investment is made by commercial companies. With this approach, the system boundaries of the CFF are kept intact, but Factor B is calculated based on the economic revenues from accepting the waste (gate fees; VGF) and energy sales (VE): B = VE/(VE+VGF) The equation is similar to economic allocation; however, with this definition of Factor B, not only burdens but also the benefits of energy recovery will be assigned to the waste treatment and recovered energy in proportion to their contribution to the total revenues of the energy-recovery plant. The approach reduces the risk of incorrect incentives. However, it does not eliminate this risk, because Factor B will depend on the economic conditions of the energy recovery, while Factor A and the quality ratio vary with the material. Ekvall et al. (2021) state that B=0.6 might be appropriate for waste incineration in Sweden, but that B is likely to be lower for incineration in countries without district-heating systems. Data on gate fees and energy revenues might not always be publicly available. When these economic data are not available, Factor B could be set at 0.5 by default. With B=0.5, the maximum error in Factor B is minimized because B can vary between 0 and 1. The risk that PEF results give wrong indication on the best waste treatment of a product or material is also lower with B=0.5, compared to B=0. However, there will still be a risk for wrong indications when A=0.2 (valid for base metals and most paper fractions) or A=0.8 (valid for textiles). The risk that PEF results point in the wrong direction is almost zero if Factor B is defined to be equal to Factor A. The problem with this approach is that Factor A is given by the conditions on the markets for recyclable materials. Linking the modelling of energy recovery to conditions on the market for recyclable materials seems far-fetched. The options that we identified for an alternative value of Factor B are presented in Table 1. They all can be justified based on different arguments. However, none of them is a perfect solution.

Tab. 1. Options for Factor B in the Circular Footprint Formula

Factor B

Justification

Drawback

Current default value in EF; corresponds to the well-established substitution approach

Can create incorrect incentives for energy recovery

VE/(VE+VGF)

Reflects the economic drivers behind investments in energy-recovery plants; reduce the risk for incorrect incentives

Not valid when investments are made without economic drivers; economic data might not be available

0.5

Minimizes maximum error in B

Very rough number; might overestimate the importance of energy demand in most countries

Almost eliminates the risk of incorrect incentives

Weak link to physical or economic realities of energy recovery

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

When we discussed this matter in the group, we observed that the incorrect incentives created by B=0 appears only in a few cases where a comparison is or can be made between recycling and energy recovery, where both options give a net benefit for the environment, and where the net benefit of recycling is greater than the net benefit of energy recovery. We have also resolved other key methodological discussions in this report by stating that the PEF methodology is our default option but that deviations are ok when they are called for. For this reason, we agreed to keep B=0 as our default value, while being aware that this is not an appropriate value in all calculations. When B=0 is not appropriate, PlastLIFE researchers should choose a more appropriate value among the options in Table 1.

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

6 Accounting for time

6.1 Prospective LCA Future-oriented LCAs go under different terms: prospective LCA, ex-ante LCA, anticipatory LCA, exploratory LCA, etc. Arvidsson et al. (2024) searched for the first three options in Scopus and found that prospective LCA (PLCA) has become a much more common term than ex-ante or anticipatory LCA. They recommend the use of PLCA for future-oriented LCA. I agree with this recommendation. In addition, I recommend using the term ex-ante LCA for assessments that are made in advance of a decision or action. This includes LCAs made to assess a policy instrument before the policy decision is made. It also includes LCAs made in product development to assess, e.g., a design or choice of material before the product is produced. An ex-ante LCA is a PLCA if it accounts for future developments in the foreground and background system. An ex-ante LCA is not a PLCA if it is based on input data that reflect the currently used technologies. One of the challenges in PLCA is that input data on new processes and the production of new materials might be available only from pilot plants, from laboratories, or from theoretical calculations. This makes it difficult to assess new materials and processes. Emerging technologies are often characterized based on their technology readiness level (TRL) from 1 to 9. The definition of each level varies slightly between sectors and actors, but the TRL of an emerging technology increases as the technology becomes more mature. The Work Programme for 2014-2015 of the EU HORIZON 2020 includes the following TRL explanation (EC 2013): 1. Basic principles observed 2. Technology concept formulated 3. Experimental proof of concept 4. Technology validated in lab 5. Technology validated in relevant environment 6. Technology demonstrated in relevant environment 7. System prototype demonstration in operational environment 8. System complete and qualified 9. Actual system proven in operational environment When a new material or process has reached full commercial production, the input data can still make the assessment biased against the new material or process, because new technologies proven in operational environment typically have a big improvement potential in energy efficiency, etc. Hence, to make a fair environmental assessment of new technologies it is often not sufficient to insert available input data in the calculations. To avoid bias against new materials or processes, a projection should be made to obtain data that reflect the environmental performance of the technology in a future when it is more mature. Several research groups have addressed this challenge (Steubing et al. 2023). Piccinno et al. (2016) present an early approach to help LCA practitioners scale up chemical production processes when only data from laboratory experiments are available. They identify and simplify the most important calculations for the energy use of liquid phase batch reactions and for certain purification and isolation steps.

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

For other input and output values, they give quantitative estimates and qualitative considerations that are needed to model the full-scale process. The approach of Imtiaz (2024) is one of the most recent contributions to this field. He uses thermodynamic process modeling and simulation for upscaling of novel chemical processes. Imtiaz (2024) applies the approach by developing a model of a scaled-up new process for carbon capture and utilization (CCU). When data on the foreground system reflect a future performance, a consistent assessment requires that the background system is also modelled with prospective data. This can be important particularly for parts of the background system that are in transition. One example is the electricity-supply system, which is in transition towards a greater share of renewable energy sources. Another example is orebased steel production, where fossil-free technologies are introduced to replace coke-fired blast furnaces. Prospective data on the background system can be important also for well-established products with a long service life, because they will have part of their use phase and the final waste management far into the future. The environmental impacts of future background processes are inherently uncertain. An LCA practitioner needs to decide how to manage this uncertainty in the assessment methodology. It is a frequently occurring challenge in LCA, which means many researchers before us had to decide on the matter. Several methods have been proposed. These includes a range of energy-systems models such as MARKAL, PRIMES, and TIMES (Plazas-Niño et al. 2022). Models that produce prospective energy data can have an even broader scope. REMIND is a multiregional general equilibrium model of the global macro-economic system with detailed representation of the energy system (Leimbach et al. 2010; Baumstark et al. 2021). It generates data on plausible developments of the energy system in different socio-techno-economic scenarios. IMAGE is a multiregional integrated assessment model that includes aspects of economic equilibrium modelling but also modelling of the environmental systems and environmental impacts (Stehfest et al. 2014). REMIND and IMAGE are both used in the development of the tool premise (PRospective EnvironMental Impact asSEment), which is an open-source Python library. premise includes a database for prospective LCA by substituting energy data in energy-intensive activities in the ecoinvent database (Sacci et al. 2022). The current PEF guidelines give no guidance on PLCA. Consistently, the default approach in PlastLIFE is to model the product system as a snapshot in time with current data in all parts. However, the prospective approaches and tools discussed above can be used when this is important for the conclusions of the assessment. The premise database might be a particularly valuable tool in this context. The reviews of Steubing et al. (2023) and Arvidsson et al. (2024) list many other potentially useful sources.

6.2 Temporal storage and release Materials that are both biobased and biodegradable (wood, paper, and certain plastics) store carbon for a limited amount of time. Temporal storage of carbon can reduce the speed of climate change and improve the chances of avoiding tipping points where climate change reinforces itself. Cement has a corresponding drawback: it generates CO2 emissions in the production stage but capture part of these emissions over a long period of time in and after the use phase of the product. Several approaches to account for temporal storage or temporal release of carbon have been suggested in the scientific literature (e.g., O’Hare et al. 2009; Levasseur et al. 2010; Röyne et al. 2016; Cowi et al. 2021) and applied in LCA and carbon-footprint calculations (e.g., Hammar & Levihn 2020; Zheng et al. 2021; Pires et al. 2024). Comprehensive reviews of available approaches have been published by, for example, Levasseur et al. (2016) and Brandaõ et al. (2019).

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

The current PEF methodology does not account for such dynamics in the carbon flows. When PlastLIFE solutions involve degradable plastics, the researchers need to decide whether to adhere to the PEF methodology on this point, or instead to account for carbon-flow dynamics. The same holds if PlastLIFE solutions affect the use of biogenic materials, biogenic fuels, or cement. Dynamic carbon modelling requires that the LCI data include the year of each emission. This makes the assessment more complex. Still, deviating from the PEF methodology can be considered justified for two reasons: • Impacts of climate change: the assessment should give an incentive for delaying emissions, because this will slow down climate change. The negative impacts of climate change are not just caused by the magnitude of this change but also by the rate of the climate change. • Consistent methodology: the climate assessment should have a consistent boundary in time. If this boundary is 100 years from now, we should account for 100 years of radiative forcing for emissions that occur now, but only for 50 years of radiative forcing for emissions that occur 50 years into the future (see Figure 3).

Fig 3. Cumulative radiative forcing of 1 tonne CO2 emitted today vs. 50 years into the future. Source: Woodworks 2025. The dynamic LCA approaches proposed by, e.g., O’Hare et al. (2009) and Levasseur et al. (2010) involves calculating the cumulative radiative forcing up to a fixed time horizon, as illustrated in Figure 3. This makes the time horizon consistent and allows for accounting for the benefit of delaying emissions and for assessing temporal storage and release of carbon. A limitation of this approach is that it accounts for the full radiative forcing up to the time horizon but completely disregards radiative forcing that occurs after this horizon. This would be fully appropriate only if radiative forcing is environmentally relevant until the time horizon, but then suddenly stops being environmentally relevant. Levasseur et al. (2019) acknowledge this limitation and compare it to discounting of future environmental impacts. They refer to Hellweg et al. (2003) who argue against discounting in LCA because 1) the environmental system is not growing like the economy but can instead shrink due to environmental degradation, and 2) the well-being of future generations should not be discounted. They also investigate and illustrate the significance of discounting by applying an annual discount factor 2% within their time horizon in some of their calculations.

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

The limitation of the fixed time horizon can be eliminated by replacing time horizon with a discounting factor on the radiative forcing. Ekvall (2017) suggests that climate impact is calculated with an accounted Global Warming Potential – aGWP(t,r) –, which is a function of the time of emission (t) and the discount rate (r). Respecting the arguments of Hellweg et al. (2003), such a discount rate should not be based on a notion that future environmental impacts are less important. Instead, the discount factor should be based on the argument that future radiative forcing is likely to have less negative impact on the environment. Such an argument can be made on several grounds: • Some climate scientists argue that the climate approaches a tipping point, where climate change is no longer reversible but will reinforce itself. If this is correct, reducing radiative forcing in the near future is particularly important. • Climate-related impacts on ecosystems and society are not only caused by the magnitude of climate change, but by the rate of climate change. If we are successful in reducing the speed of climate change, the rate of change is at its peak now or in the near future. Reducing near-future radiative forcing is particularly important to reduce and shorten this peak. • If the climate reaches a tipping point, it will change rapidly until it reaches a new stable state. After this potentially catastrophic development, the rate of climate change is likely to be lower. Human-induced radiative forcing might for this reason have less impact on the ecosystems and society that remains at this stage. • Regardless of tipping points, climate change is already affecting both ecosystems, agricultural systems, and society at large. When ecosystems and society have adapted to a new, warmer climate, rapid cooling of the climate is likely to have new negative impacts on the adapted systems. In this future, human-induced radiative forcing will contribute to upholding the warmer climate. Hence, future human-induced radiative forcing might have positive environmental impacts. With a high discount rate, the aGWP declines rapidly over time. A delay in emissions or a temporal carbon storage has a large impact on the aGWP: if r=5% per year (cf. Figure 4), emissions of CO2 that occur 50 years into the future are tentatively assigned just 6% of the aGWP assigned to the same quantity of CO2 emitted today. With a high discount rate, the aGWP also assigns more weight to short-lived greenhouse gases (GHGs) such as methane, compared to CO2; this is similar to traditional GWP with a short time horizon.

Fig 4. The decline over time of CO2 emitted now or 50 years into the future, and its accounted radiative forcing with discount rates of 0.2% and 5% per year. Source: Ekvall (2017).

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

With a low discount rate, the aGWP declines more slowly over time. This indicates a small or moderate benefit from a delay in emissions or temporal carbon storage: if r=0.2% per year (Figure 4), remissions of CO2 that occur 50 years into the future are tentatively assigned 83% of the aGWP assigned to the same quantity of CO2 emitted today. A low discount rate also assigns more weight to CO2 and less to short-lived GHGs; in this sense it resembles traditional GWP with a long time horizon. The choice of discount rate is subjective, just like the choice of time horizon in traditional GWP calculations. The strong emphasis on near-future radiative forcing given by a high discount rate might be relevant for decision-makers that focus on avoiding tipping points in the climate. A low discount rate might be more relevant for decision-makers where the main argument for the discount rate is that human-induced radiative forcing might be beneficial far into the future. Parallel calculations with a high and a low discount rate can be made to give guidance to both kinds of decision-makers. Calculating cumulative radiative forcing with or without a discount rate requires a bit of complex mathematics. When the necessary tools and expertise is not available, the simplified approach of Clift & Brandaõ (2008) can be an attractive option. They propose to reduce the GWP by 0.76% for each year of delay in GHG emissions, if this delay is no more than 25 years. They argue that this is a good approximation of the results obtained when calculating the cumulative radiative forcing with a fixed time horizon of 100 years. The British standard for carbon-footprint calculations (PAS 2050; BSI 2011) requires that the approach of Clift & Brandaõ (2008) be used for assessing GHG emissions from the use phase or the final disposal phase of a product that occur as a single release within 25 years of the production of the product. In the general case of delays in emissions, PAS 2050 instead requires that the GWP is reduced by 1% for each year of delay in GHG emissions. This approach is even simpler and cruder and, in addition, inconsistent with the approach of Clift & Brandaõ (2008). Noting the multitude of approaches available when the current PEF guidelines were written, the European Commission decided that the methodology for dynamic modelling of GHG emissions was too immature. For this reason, the current PEF Guidelines advise against dynamic carbon modelling. The guidelines are now being revised, and dynamic carbon modelling might be included in the revised guidelines. Based on the above, I recommend, for now, that PlastLIFE researchers apply static carbon modelling in the general case, while being aware that this approach disregards temporal storage, temporal release, and delayed release of carbon. When these dynamic features are important for the conclusions of the assessment, an approach for dynamic carbon modelling should be considered. It would be interesting to further develop the discounting approach (Ekvall 2017), and to test it in case studies. In PlastLIFE, dynamic carbon modelling can be relevant in the case study on reuse, which is planned for 2026. It might also be relevant as an extension of the national model, for estimating and discussing the implications of using plastics in constructions.

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

7 Electricity modelling

When electricity modelling is discussed, a distinction is often made between the market-based and location-based approaches. Market-based electricity modelling means that data on specific technologies for electricity supply should be used when a valid contractual instrument (Guarantees of Origin, Power-Purchasing Agreements etc.) states this technology as the source of the electricity. When the electricity is bought without such a contractual instrument, the electricity supply should be modelled using data that represents the residual mix, i.e., the part of the electricity mix that is not sold with the contractual instruments. Market-based electricity modelling is on the surface attractive in ALCA, since it allows for the use of specific information when such information is available. At the same time, it avoids double counting of, for example, renewable electricity production. It also gives a competitive edge to electricity suppliers that generate renewable electricity, which gives an incentive for investments in renewable power production. The market-based approach has problems, though (Ekvall et al. 2023). The approach is difficult to apply accurately, particularly for electricity production outside Europe, because it is difficult to check if a contractual instrument is valid and because data on residual electricity are unavailable for large parts of the world. The approach can allow for greenwashing, because the connection between a contractual instrument (e.g., Guarantees of Origin) and an increase in the production of renewable electricity can be weak or non-existent. The connection between contractual instruments and the physical reality can also be weak: electricity users on the European continent can, for example, buy Guarantees of Origin that state that they use geothermal electricity from Iceland, although there is no grid connection between Iceland and continental Europe. Location-based electricity modelling, in contrast, does not account for contractual instruments. Any electricity use is modelled based on average data for a country or region, regardless of Guarantees of Origin or other contractual instruments. The location-based approach can also be attractive in attributional LCA. It is consistent with the focus of attributional LCA to model the physical reality and disregard market mechanisms. The most common location-based approach is to use data that represent a national electricity mix. However, Ekvall et al. (2023) argue that this overestimates the significance of national boundaries in the assessment, since the electricity grid has developed into an international system. The physical reality would be better reflected by average data over a larger geographical electricity grid. The international trade in electricity calls for a distinction between the production mix and the consumption mix. The production mix refers to the electricity produced in a country or region; the consumption mix refers to the electricity used in the country or region. The difference is that the latter includes imported electricity. The distinction between production and consumption mix should be made when modelling the residual electricity mix in a market-based approach, and when modelling the average mix in a location-based approach. This increases transparency and conceptual clarity in the LCA. However, the difference in results is likely to be small, because imported electricity is a small share of the total electricity mix in most countries. All the above is applicable for attributional LCA, where electricity data reflect an attempt to identify how the electricity is produced. In consequential LCA, in contrast, electricity data ideally reflect how

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

the electricity use affects electricity production. Marginal data are often used for this purpose. Here, it is useful to distinguish between short-term and long-term marginal data. Short-term marginal data reflect how the use of existing power plants is affected by a change in electricity use, until the production capacity has had time to adapt to this change. Long-term marginal data reflect the changes in production capacity that are made in response to the change in electricity use. Short-term marginal data can be obtained through models of the existing electricity system. They can also be developed based on the electricity prices, because the electricity price ideally reflects the running cost of the short-term marginal supply. Long-term marginal data are more difficult to develop. It is not possible to know what construction or closure of power plants are affected by the electricity use in a specific product system. Hence, the real long-term marginal impacts cannot be identified. When we use long-term marginal data, these are only valid given a specific method or model. Available methods range from simple assumptions to advanced energy-systems modelling. Each method has its own limitations. The PEF rules stipulate that electricity be modelled with a market-based approach. When the electricity is bought without a contractual instrument, the electricity supply should be modelled using data that represents the national residual consumption mix, i.e., the part of the electricity use in the country that is not specified through valid contractual instruments. For European electricity, the PEF rules refer to data on from the Association of Issuing Bodies (AIB). After discussing the problems of the market-based approach and the use of national average data, Ekvall et al. (2023) propose two other approaches: • In attributional LCA, location-based electricity data can be developed that represent the average production in the effective electricity market where the power is used. • In consequential LCA, the marginal data from the Ecoinvent database can be used. The PlastLIFE group did not consider the market-based approach problematic as such. The incentive it gives to investments in renewable energy was highlighted as a benefit. However, it might be relevant to add conditions on when the market-based approach is valid. One condition can be that contractual instruments are only accounted for when there is a physical grid connection that makes the transfer of electricity possible. Another condition can be evidence of additionality. An example of such evidence is that the electricity is used when the electricity price is very low. The low electricity price indicates that wind and solar power produces an excess of electricity, and electricity use at such time means that more of the solar and wind power is utilized. If this condition is introduced, data on electricity use will need a great temporal resolution – for example through hourly measurements. Regardless of what requirement is added to the market-based approach, more electricity will be included in the residual mix. This means that residual data from the AIB will no longer be accurate. Developing alternative residual data will be a challenge, because it requires information (or assumptions) on what electricity use that meets the requirement in the whole country or region. The PlastLIFE group noted that what electricity approach is appropriate can vary from case to case, depending on the goal and scope of the study. However, it is not easy to set rules for when market-based and location-based approaches should be used. The rule that can be given is that both approaches can be applicable in attributional LCA, while marginal data are more applicable in consequential LCA.

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

8 Mass-balance and book-and-claim

Mass-balance and book-and-claim approaches are examples of chain-of-custody (CoC) models. ISO 22095 defines chain of custody as “a process by which inputs and outputs and associated information are transferred, monitored and controlled as they move through each step in the relevant supply chain” (ISO 2020b). Relevant information to pass on through the supply chain in the PlastLIFE project includes the share of biobased or recycled material. The CoC models include other approaches, where the rules for passing on information are stricter. The full list, as presented by ISO (2020b) is (cf. Figure 5): • Identity Preserved: the material flow from each source is kept physically separated. • Segregated: the material flow is from a mix of sources, but material of each kind is kept physically separated. • Controlled blending: the material flow is a mix of types in known and constant proportions. • Mass balance: the material flow is a mix of types in varying proportions. • Rolling average mass balance: the producer reports average values for a mix that varies over time • Credit-based mass balance: the producer claims that the biobased or recycled material is concentrated in part of the products even when the actual mix of material is the same in all products. • Book and claim: suppliers of biobased or recycled material can sell the material to one customer and the label “biobased” or “recycled” to another customer. The labelled material need not contain any share of the stated kind of material.

Fig 5. Chain-of-custody models. Source: ISO 2020b.

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

The first three models and rolling-average mass balance are all uncontroversial in LCA. Credit-based mass balance and book-and-claim approaches are more controversial. Renewable Carbon Initiative (2022) states that these approaches are frequently applied. They argue that this is important for the chemical industry to be able to transition from fossil to biobased and/or recycled raw materials. Finkbeiner et al. (2025), on the other hand, argue that credit-based mass balance and book-andclaim approaches: 1. are inconsistent with the LCA approach that focusses on modelling physical flows, since the flows that are modelled with these approaches can be very different from the physical flows; 2. lead to inconsistency between the accounting methods for products, organisations, and countries; 3. can lead to unjust comparisons between products that are modelled with differing methods; 4. give a false or at least unproven impression that the use of a product leads to an increased supply of recycled or biobased material; and 5. are inconsistent with the allocation hierarchy in ISO 14044. Some of these arguments are stronger than others. None of the fully disqualifies credit-based mass balance or book-and-claim approaches: Re 1: this argument is valid from the point of view that LCA practitioners should focus on modelling the physical reality and disregard whether the LCA generates information that is relevant for decision-making or creates good incentives. This point of view excludes not only mass balance and bookand-claim but also, for example, consequential LCA and prospective LCA. Finkbeiner et al. (2025) suggests that these should all be called “message LCA” rather than just LCA. Re 2: this argument is probably correct. However, there will probably always be inconsistencies in LCA and accounting methods that are used in different contexts. Re 3: this argument is correct. When different products are compared, it is essential that the methods are the same. This holds not just for CoC models but for all significant methodological choices in the LCA. Re 4: this is a straw-man’s argument. Finkbeiner et al. (2025) observe that the mass-balance and book-and-claim approaches are defended with the argument that they support the transformation of industry. They then argue that this defense is valid only if it can be proved that the use of biobased/recycled material leads to a corresponding increase in the supply. However, there is no need to prove actual additionality to argue that mass-balance and book-and-claim approaches are likely to create an incentive for increasing the supply of biobased/recycled material. It is possible but not necessary to require additionality when mass-balance and book-and-claim approaches are applied. Re 5: Finkbeiner et al. (2025) argue that the mass-balance and book-and-claim approaches create an allocation problem because the process generates multiple products that differ in terms of assigned biobased/recycled content. Since they are produced in a joint process, allocation should be based on mass or economic value. The raw-material use of the production process is normally included in the allocation: the biobased/recycled material is allocated to each product based on the mass or economic value. Since the same allocation factors are applied to the fossil-based material, the mix of input material will be the same for all products. Hence, the credit-based mass balance is made invalid. A possible solution to this is to exclude the raw-material input from allocation problems caused by CoC models. The allocation is then reduced to include only the energy use, ancillary materials, emissions, and waste generated by the production process. This discussion is related to the debate on location-based and market-based electricity (Ekvall et al. 2025). Location-based electricity is similar to the rolling-average mass balance, because national average electricity is typically given per year, while the actual electricity supply mix varies over the day and between seasons. Market-based electricity can be considered a case of book and claim. Because power suppliers can sell the electricity to one customer and the Guarantees of Origin to another customer.

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

In the debate on book-and-claim approaches, there is a potentially important difference between electricity and materials: residual data are available for market-based electricity in Europe, but not for materials. This makes it more difficult to properly apply the book-and-claim approaches for material. Further discussion is warranted on the similarities and differences between CoC models for electricity and materials. This could be a suitable topic for a paper. The current PEF guidelines stipulate book-and-claim for electricity, but do not encourage the use of mass-balance and book-and-claim approaches otherwise. As part of the ongoing revision of the guidelines, the issue was discussed with the EF Technical Advisory Board in March 2025. The Board was divided. The PlastLIFE group found the book-and-claim approach problematic for materials, because it invites greenwashing and cherry-picking. However, some of the characteristics were still considered to support well-functioning secondary markets and potentially reduce unnecessary physical transports, which would benefit the environment. Based on the above, I recommend that the book-and-claim approach should not be applied to materials in PlastLIFE. Mass-balance approaches can be applied, but this should be transparently reported and justified.

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

9 Outlook

It is difficult and probably counterproductive to prescribe what methods PlastLIFE researchers should apply. This report leaves much room for methodological choices that the researchers deem relevant. In most cases, however, the PEF framework can be considered the default approach and deviations from this framework should be transparently reported and justified. It should be noted that the PEF framework is currently being revised. A new version is likely to be adopted in 2027, i.e., before the PlastLIFE project ends. This could affect the default methods used in case studies during the final parts of the project. In the historic and current debate on LCI methodology, there seem to be two camps or extremes. One group of LCA experts, represented by, e.g., Finkbeiner et al. (2025), have the implicit or explicitstarting point that LCA should model the physical reality as closely as possible. The focus is on physical flows, making the life cycle an extension of the flow charts in the process industry. Experts in this group typically prefer attributional LCA over consequential LCA; they are also sceptic towards massbalance with credits and book-and-claim. It is important that input data and results are precise. The other group have the implicit or explicit starting point that LCA should have a positive impact on the environment. It should generate information that is relevant to decision-makers and generate incentives for good decisions. Researchers in this group can regard the life cycle as a large sociotechnical system, where physical flows, economic transactions, and human decision-makers interact. They can prefer consequential LCA and/or find a use for credit-based mass balance and book-and-claim approaches. The conflicting starting points of these groups of LCA experts can explain why conflicts over LCI methodology are difficult to resolve. Making LCA experts aware of this explanation might increase mutual understanding, reduce the conflict level, and, hence, increase the chances of constructive debates. LCA commissioners would also benefit from understanding this fundamental conflict in the perception of the purpose and nature of LCA. This would be an interesting direction for further work in PlastLIFE and/or beyond. The specific key topics in this report also warrant further work in and/or beyond PlastLIFE: assessment of littering and microplastics, several aspects of the CFF, prospective LCA, assessment of temporal storage and release of carbon, the pros and cons of market-based electricity, and their relations to the current debate on credit-based mass balance and book-and-claim approaches. In particular, it would be interesting to: • investigate when the CFF can be replaced by a simpler method without loss of important information; • calculate Factor B in the CFF based on the equation B = VE/(VE+VGF), suggested by Ekvall et al. (2021); • try using the premise database for prospective modelling of future processes; • further develop and test the discounting approach proposed by Ekvall (2017) for modelling delayed GHG emissions; and • discuss deeper the similarities and differences between CoC models for electricity and materials.

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

We might also need to discuss additional key issues. One potential key issue that is not covered by this report is how to model heat supply. The modelling of district-heat supply can be important for the results because the CFF includes the energy supply substituted through waste incineration with energy recovery. Heat is also used as an input to, for example, drying processes.

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

References

Arvidsson, R., Svanström, M., Sandén, B.A., Thonemann, N., Steubing, B. & Cucurachi, S. 2024. Terminology for future‑oriented life cycle assessment: review and recommendations. Int. J. LCA 29:607–613 https://doi.org/10.1007/s11367-023-02265-8 Baumstark, L., Bauer, N., Benke, F., et al. 2021. REMIND2.1: transformation and innovation dynamics of the energy-eco nomic system within climate and sustainability limits. Geosci. Model Dev. 14:6571–6603. https://doi.org/10.5194/gmd-14-6571-2021 Brandão, M., Kirschbaum, M.U.F, Cowie, A.L. & Hjuler, S.V. 2019. Quantifying the climate change effects of bioenergy systems: Comparison of 15 impact assessment methods. GCB Bioenergy 11:727–743. https://doi.org/10.1111/gcbb.12593 Brodén, E. 2022. Granulat från konstgräsplaner till havet. Kommunernas Internationella Miljöorganisation. url: KIMO Sweden's latest report looks at microplastic pollution from artificial turf – KIMO (in Swedish). BSI. 2011. PAS 2050:2011: Specification for the assessment of the life cycle greenhouse gas emissions of goods and services. British Standards Institution. Ciroth, A. & Kouame, N. 2019. Elementary litter in life cycle inventories, approach and application. Presentation at the Life Cycle Management (LCM2019) conference, Poznan, 1-4th September 2019. Cited by Nessi et al. (2021). Clift, R. & Brandaõ, M. 2008. Carbon storage and timing of emissions. Centre for Environmental Strategy Working Paper 02/08. University of Surrey, Guildford, UK. url: Microsoft Word - 0208 CES WP Carbon storage and timing of emissions.doc Corella-Puertas, E., Guieu, P., Aufoujal, A., Bulle, C. & Boulay A-M. 2022. Development of simplified characterization factors for the assessment of expanded polystyrene and tire wear microplastic emissions applied in a food container life cycle assessment. J. Ind. Ecol. 26:1882–1894. https://doi.org/10.1111/jiec.13269. Corella-Puertas, E., Carla Hajjar, C., Jérôme Lavoie, J. & Boulay, A.-M. 2023. MarILCA characterization factors for microplastic impacts in life cycle assessment: Physical effects on biota from emissions to aquatic environments. J. Cleaner Prod. 418: 138197. https://doi.org/10.1016/j.jclepro.2023.138197. Cowie, A.L, Berndes, G., Bentsen, N.S., Brandão, M., Cherubini, F., Egnell, G., George, B., Gustavsson, L., Hanewinkel, M., Harris, Z.M., Johnsson, F., Junginger, M., Kline, K.L., Koponen, K., Koppejan, J., Kraxner, F., Lamers, P., Majer, S., Marland, E., Nabuurs, G.-J., Pelkmans, L., Sathre, R., Schaub, M., Smith Jr., C.T., Soimakallio, S., Van Der Hilst, F., Woods, J. & Ximenes, F.A. 2021. Applying a science-based systems perspective to dispel misconceptions about climate effects of forest bioenergy. GCB Bioenergy 13:1210–1231. https://doi.org/10.1111/gcbb.12844

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

EC. 2013. Annex G in HORIZON 2020 WORK PROGRAMME 2014 – 2015 18. General Annexes. https://www.h2020.md/sites/h2020/files/1595112-18._general_annexes_wp20142015_rev1_en.pdf EC. 2021. COMMISSION RECOMMENDATION (EU) 2021/2279 of 15 December 2021 on the use of the Environmental Footprint methods to measure and communicate the life cycle environmental performance of products and organisations. Official Journal of the European Union L 471:16&204-593. Edwards, C. & Parker, G. 2012. A Life Cycle Assessment of Oxo-biodegradable, Compostable and Conventional Bags. Interetek Expert Services Report on behalf of Symphony Environmental Ltd. 46pp. Ekvall, T. 2017. The time horizon in assessment of climate impacts. Presentation at IEA Bioenergy Task 38 workshop “Understanding the Climate Effects of Bioenergy Systems”, May 16th 2017, Gothenburg, Sweden. url: Gothenburg, Sweden 2017. Ekvall, T. 2023. Is the distinction between attributional and consequential LCA useful? Life Cycle Talks. Swedish Life Cycle Center. https://www.youtube.com/watch?v=mnG3COjTGkA. Ekvall, T., Gottfridsson, M., Nellström, M., Nilsson, J., Rydberg, M. & Rydberg T. 2021. Modelling incineration for more accurate comparisons to recycling in PEF and LCA. Waste Management 136:153-161. https://doi.org/10.1016/j.wasman.2021.09.036 Ekvall, T., Lorentzon, K., Einarson Lindvall, E. & Palander, S. 2023. Modelling electricity in environmental footprints. Report 2023:02. Swedish Life Cycle Center, Gothenburg, Sweden. Ekvall, T., Granström, L., Jansson, R., Moberg, E. & Rydberg, T. 2024. Modelling recycling with a basket of functions applied to advanced sorting of plastic packaging waste. Presentation and poster at 4th conference on Life Cycle Assessment of Waste. June 11-14, 2024. Helsingor, Denmark. Ekvall, T., Lopez Marin, N., Liu, Y., Riise, E. & Zuin, S. 2025. Electricity and the chain of custody. Poster and short oral presentation at LCM 2025, Palermo, 9-12 September 2025. EU. 2025. Microplastics: sources, effects and solutions. url: Microplastics: sources, effects and solutions | Topics | European Parliament. Accessed July 31st 2025. ExcelPlas Australia. 2004. The impacts of degradable plastic bags in Australia. Centre for Design at RMIT University, Melbourne, Victoria. Finkbeiner, M., Roche, L., Holzapfel, P. 2025. From analysis-LCA to message-LCA: a lost cause? Int. J. LCA 30:803-810. https://doi.org/10.1007/s11367-025-02460-9 Frias, J.P.G.L. & Nash, R. 2019. Microplastics: Finding a consensus on the definition. Marine Pollution Bulletin, 138, 145–147. https://doi.org/10.1016/j.marpolbul.2018.11.022. Referred to by Lavoie et al. (2021). Høiberg, M.A., Woods, J.S. & Verones, F. 2022. Global distribution of potential impact hotspots for marine plastic debris entanglement. Ecological Indicators 135:108509. https://doi.org/10.1016/j.ecolind.2021.108509. Hammar, T. & Levihn, F. 2020. Time-dependent climate impact of biomass use in a fourth generation district heating system, including BECCS. Biomass & Bioenergy 138:105606. https://doi.org/10.1016/j.biombioe.2020.105606. Hellweg, S., Hofstetter, T.B. & Hungerbühler, K. 2003. Discounting and the environment - Should current impacts be weighted differently than impacts harming future generations. Int. J. LCA 8 (1):8– 18.

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

Horn, R., Hong, S.H., Knüpffer, E., Alvarenga, R., Boone, L., Preat, N., Behm, K., Pajula, T. &, Pihkola, H. 2021. Critical evaluation of environmental approaches. D1.1. ORIENTING. Available from ORIENTING-D1.1-Critical evaluation of environmental approaches [Accessed: September 11th 2023] Imtiaz, H.F. 2024. A Framework for Upscaling of Emerging Chemical Processes Based on Thermodynamic Process Modeling and Simulation. ChemEngineering 8:46. https://doi.org/10.3390/chemengineering8030046 ISO. 2020a. Environmental Management – Life Cycle Assessment – Requirements and guidelines: Amendment 2. ISO 14044:2006 AM2:2020. International Organisation for Standardization, Geneva, Switzerland. ISO. 2020b. Chain of custody — General terminology and models. ISO22095:2020. International Organisation for Standardization, Geneva, Switzerland. Landrigan, P.J., Dunlop, S., Treskova, M., et al. 2025. The Lancet Countdown on health and plastics. The Lancet online first, August 03, 2025. url: The Lancet Countdown on health and plastics - The Lancet. Lavoie, J., Boulay, A.-M. & Bulle, C. 2021. Aquatic micro- and nano-plastics in life cycle assessment: development of an effect factor for the quantification of their physical impact on biota. J. Ind. Ecol. 26:2123–2135. https://doi.org/10.1111/jiec.13140. Leimbach, M., Bauer, N., Baumstark, L., Luken, M. & Edenhofer, O. 2010. Technological Change and International Trade – Insights from REMIND-R. Energ. J. 31:109-136. https://doi.org/10.5547/ISSN0195-6574-EJ-Vol31-NoSI-5 Levasseur, A., Lesage, P., Margni, M., Deschenes, L. & Samson, R. 2010. Consider Time in LCA: Dynamic LCA and its Application to Global Impact Assessments. Env. Sci. Tech. 44:3169-3174. https://doi.org/10.1021/es9030003 Levasseur, A., Cavalett, O., Fuglestvedt, J.S., Gasser, T., Johansson, D.J.A., Jörgensen, S.V., Raugei, M., Reisinger, A., Schivley, G., Strömman, A., Tanaka, K. & Cherubini, F. 2016. Enhancing life cycle impact assessment from climate science: Review of recent findings and recommendations for application to LCA. Ecological Indicators 71:163-174. https://doi.org/10.1016/j.ecolind.2016.06.049 Loubet, P., Couturier, J., Horta Arduin, R. & Sonnemann, G. 2022. Life cycle inventory of plastics losses from seafood supply chains: Methodology and application to French fish products. Science of the Total Environment 804:150117. https://doi.org/10.1016/j.scitotenv.2021.150117. Maga, D., Galafton, C., Blömer, J., Thonemann, N., Özdamar, A. & Bertling, J. 2022. Methodology to address potential impacts of plastic emissions in life cycle assessment. Int. J. LCA 27:469–491. https://doi.org/10.1007/s11367-022-02040-1. Nessi, S., Sinkko, T., Bulgheroni, C., Garcia-Gutierrez, P., Giuntoli, J., Konti, A., Sanye Mengual, E., Tonini, D.., Pant, R., Marelli, L. & Ardente, F. 2021. Life Cycle Assessment (LCA) of alternative feedstocks for plastics production - Part 1: the Plastics LCA method, EUR 30725 EN, Publications Office of the European Union, Luxembourg, ISBN 978-92-76-38145-7, https://doi.org/10.2760/271095, JRC125046 O’Hare, M., Plevin, R.J., Martin, J.I., Jones, A.D., Kendall, A. & Hopson, E. 2009. Proper accounting for time increases crop-based biofuels’ greenhouse gas deficit versus petroleum. Environ. Res. Lett. 4(2):024001. https://doi.org/10.1088/1748-9326/4/2/024001 Peano, L., Kounina, A., Magaud, V., Chalumeau, S., Zgola, M. & Boucher, J. 2020. Plastic Leak Project – Methodological Guidelines. V1.3. Quantis.

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

Piccinno, F., Hischier, R., Seeger, S. & Som, C. 2016. From laboratory to industrial scale: a scale-up framework for chemical processes in life cycle assessment studies. J. Cleaner Prod. 135:10851097. https://doi.org/10.1016/j.jclepro.2016.06.164 Pires, S.T., Williams, A., Daystar, J., Sagues, W.J., Lan, K. & Venditti, R.A. 2024. Evaluating Cotton Apparel with Dynamic Life Cycle Assessment: The Climate Benefits of Temporary Biogenic Carbon Storage. BioResources 19(3): 5074-5095. https://doi.org/10.15376/biores.19.3.5074-5095 Plazas-Niño, F.A., Ortiz-Pimiento, N.R. & Montes-Páez, E.G. 2022. National energy system optimization modelling for decarbonization pathways analysis: A systematic literature review. Renewable and Sustainable Energy Reviews 162:112406. https://doi.org/10.1016/j.rser.2022.112406 Renewable Carbon Initiative. 2022. Importance of mass balance and free attribution (MBFA) for the conversion of the chemical sector of alternative carbon sources. A position paper of the Renewable Carbon Initiative. url: Importance of mass balance and free attribution (MBFA) for the conversion of the chemical sector to alternative carbon sources (October 2022) | Renewable Carbon Publications Roosen, M., Tonini, D., Albizzati, P.F., Caro, D., Cristóbal, J., Lase, I.S., Ragaert, K., Dumoulin, A. & De Meester, S. 2023. Operational Framework to Quantify “Quality of Recycling” across Different Material Types. Env. Sci. Tech. 57(36):13669-13680. https://doi.org/10.1021/acs.est.3c03023 Röyne, F., Peñaloza, D., Sandin, G., Berlin, J. & Svanström, M. 2016. Climate impact assessment in the life cycle assessment of forest products: implications of method choice for results and decisionmaking. J. Cleaner Prod. 116:90-99. https://doi.org/10.1016/j.jclepro.2016.01.009 Sacchi, R., Terlouw, T., Siala, K., Dirnaichner, A., Bauer, C., Cox, .B, Mutel, C., Daioglou, V. & Luderer, G. 2022. PRospective EnvironMental Impact asSEment (premise): A streamlined approach to producing databases for prospective life cycle assessment using integrated assessment models. Renewable and Sustainable Energy Reviews 160:112311. https://doi.org/10.1016/j.rser.2022.112311 Salieri, B., Stoudmann, N., Hischier, R., Som, C. & Nowack, B. 2021. How Relevant Are Direct Emissions of Microplastics into Freshwater from an LCA Perspective? Sustainability 13:9922. https://doi.org/10.3390/su13179922 Schulte, A., Salinas Velarde, P.A., Marbach, L & Mörbitz, P. 2023. Measuring the circularity potential of recycled LDPE based on quantity and quality conservation - a functional requirement matrix approach. Resources, Conservation & Recycling Advances 17:200127. https://doi.org/10.1016/j.rcradv.2022.200127. Stehfest, E. et al. eds. 2014. Integrated assessment of global environmental change with IMAGE 3.0: Model description and policy applications. Netherlands Environmental Assessment Agency (PBL), Den Haag, The Netherlands. Steubing, B., Mendoza Beltran, A., Sacchi, R. 2023. Conditions for the broad application of prospective life cycle inventory databases. Int. J. LCA 28:1092-1103. https://doi.org/10.1007/s11367-02302192-8 The Economist. 2025. Microplastics have not yet earned their bad reputation. url: Microplastics have not yet earned their bad reputation. Accessed August 4th 2025. Tillman, A.-M. 2000. Significance of decision-making for LCA methodology. Environ. Impact Assessment Review 20:113-123. https://doi.org/10.1016/S0195-9255(99)00035-9. Tillman, A.-M., Ekvall, T., Baumann, H. & Rydberg, T. 1994. Choice of system boundaries in life cycle assessment. J. Cleaner Prod. 2(1):21-29. https://doi.org/10.1016/0959-6526(94)90021-3.

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

Vadenbo, C., Hellweg, S. & Astrup, T.F. 2017. Let's Be Clear(er) about Substitution: A Reporting Framework to Account for Product Displacement in Life Cycle Assessment. J. Ind. Ecol. 21:10781089. https://doi.org/10.1111/jiec.12519. Weidema, B.P. 2003. Market Information in life cycle assessment. Environmental Project no. 863. Copenhagen: Danish Environmental Protection Agency. Available from: https://www2.mst.dk/Udgiv/publications/2003/87-7972-991-6/pdf/87-7972-992-4.pdf [Accessed: August 25th 2023]. Woods, J.S., Verones, F., Jolliet, O., Vazquez-Rowe, I. & Boulay, A.-M. 2021. A framework for the assessment of marine litter impacts in life cycle impact assessment. Ecological Indicators 129:107918. https://doi.org/10.1016/j.ecolind.2021.107918. Woodworks. 2025. Long-Term Biogenic Carbon Storage. Available from: https://www.woodworks.org/resources/long-term-biogenic-carbon-storage/ [Accessed: May 22nd 2025]. Zheng, D., Cao, H., Ma, C., Hauschild, M.Z. & Dong, Y. 2021. A dynamic approach for life cycle global warming impact assessment of machine tool considering time effect. Int. J. LCA 26:1391– 1402. https://doi.org/10.1007/s11367-021-01933-x.

Report of the PlastLIFE project | Methodological considerations in plastics-related life cycle assessment

Report of the PlastLIFE project coordinated by the Finnish Environment Institute:

Methodological considerations in plastics-related life cycle assessment ISBN 978-952-11-5866-7 (online) Finnish Environment Institute, 2026.