Incorporating Recyclates Derived from Household Waste into Flexible Food Packaging Applications: An Environmental Sustainability Assessment

Abstract

Integrating recyclates into food packaging is key towards circularity while meeting functionality and safety requirements; however, associated environmental impacts remain underexplored. This gap was addressed through a cradle-to-gate life cycle assessment, using the Environmental Footprint method, along with substitution and cut-off approaches for handling the multifunctionality of recycling. Recyclates were derived from polyethylene (PE)-rich household food packaging waste, purified via delamination-deinking. Firstly, results show that shifting from virgin multi-material to mono-material multilayer structures with or without recyclates, while maintaining functionality, offers environmental benefits. Secondly, recyclates should sufficiently substitute virgin materials in quantity and quality, decreasing the need for primary plastics and avoiding recyclate incorporation without functionality. Otherwise, thicker laminates are obtained, increasing processability challenges and environmental impacts, e.g., 12% for particulate matter, and 14% for mineral-metal resource use when the recycle content rises from 34 to 50%. Thirdly, a fully closed loop for flexible food packaging is not yet feasible. Key improvements lie in reducing residues generated during recycling, especially in delamination-deinking, lowering energy use in recompounding, and using more efficient transport modes for waste collection. Further research is essential to optimise the innovative technologies studied for flexible food packaging and refine them for broader applications.

1. Introduction

Packaging is the largest end-use sector for plastics in Europe, accounting for 39% of total demand in 2022, followed by building and construction at 23%. However, recycled plastic content in packaging remains limited, with only 15.9% incorporated in 2022, primarily via mechanical recycling (9.5% from post-consumer and 6.2% from pre-consumer plastic waste) [1]. Recycled content refers to the proportion by mass of recycled material in a product. For packaging, only pre-consumer and post-consumer materials are considered as recycled content [2]. This heavy reliance on virgin plastics raises environmental concerns [3]. To address these issues, the European Union (EU) has established regulations to enhance packaging recyclability and foster a circular economy. The Packaging and Packaging Waste Regulation (EU) 2025/40, effective from February 2025 and fully applied from August 2026 with transitional dates through 2040, establishes comprehensive requirements for packaging within the EU, e.g., mandating minimum levels of recycled plastic content, setting targets to reduce packaging waste and enhance recycling rates [4]. For instance, contact-sensitive packaging made of plastics other than polyethylene terephthalate (PET), except single-use beverage bottles, shall contain the minimum recycled content, recovered from post-consumer plastic waste, of 10% by 2030 and 25% by 2040. Complementing this, Regulation (EU) 2025/351 [5], effective from March 2025, sets strict requirements for the use of plastic materials and articles intended to come into contact with food and minimum requirements for a quality assurance system to be operated at recycling facilities, where recycled plastic is manufactured following Regulation (EU) 2022/1616 [6]. Retaining treatment of plastic and packaging waste within the EU towards circular economy, in addition to investment in EU-based recycling infrastructure, is also encouraged by the European Parliament and the Council through the Regulation (EU) 2024/1157, effective from May 2024 and fully applied from May 2026 [7].

Flexible packaging, of which use is becoming increasingly widespread, is also impacted by these regulations. It encompasses monolayer films for tertiary and secondary packaging, as well as multilayer films for primary food packaging [8]. Multilayer films are categorised into (i) multi-material multilayer (MuMu), composed of diverse polymeric and non-polymeric materials (e.g., aluminium, paper), and (ii) mono-material multilayer (MoMu), composed of a single polymer [9]. The most commonly used polymers in multilayer flexible packaging are polyethylene (PE), polypropylene (PP), PET, polyamides (PA or nylon), followed by ethylene-vinyl alcohol (EVOH), and ethylene vinyl acetate (EVA) [10]. MuMu structures dominate in packaging for food and sensitive non-food products that require chemical barrier properties [11]; however, they pose considerable recycling challenges [12]. In contrast, multilayer films containing PE or PP with a minimum content of 90 wt% are considered as recyclable MoMu, which is seen as a beneficial structure for recycling [10]. However, the market penetration of their recyclates remained low so far, despite their largest share in post-consumer plastic waste [13].

To address the recycling limitations of multilayer packaging, advanced technologies have emerged. Focusing on sorting, different solutions have been developed for tracking plastics, categorising them into four groups: tracer (or marker), blockchain, digital product passport, and certified system. Among them, the first group has attracted the attention of both academia and industry. Several companies investigated physical tracking by tracers, e.g., Polysecure [14]. Moreover, chemical recycling is considered a complement to mechanical recycling, enabling the recycling of mixed post-consumer plastic waste by changing its chemical structure to recyclates with high quality suitable for contact-sensitive applications (e.g., in health and food packaging). However, its share is currently only 0.1% of European plastics production, with the infrastructure mainly built for depolymerisation and pyrolysis technologies [1]. Plastic purification through selective dissolution or delamination also shows potential. Selective dissolution, referred to as solvent-based dissolution, solvent-based purification or dissolution-precipitation, enables targeted polymer recovery based on their differential solubility in specific solvents, while delamination removes adhesives between layers to yield purer recyclates [9]. Today, these technologies remain largely in the research and pilot phase [12] and face economic barriers to scale-up [15]. Industrial-scale implementation remains limited. A notable exception is Saperatec’s initiative in 2022 to construct a recycling facility with an annual capacity of 17,000 tons, aiming to process flexible multilayer waste (including both MuMu and MoMu) through delamination [16]. Simultaneously, industry actors are making progress in incorporating recyclates into flexible laminates. For example, Ineos has introduced a recyclable MoMu film containing over 50% recycled content in 2023, although its specific end-use applications (food vs. non-food) have not been disclosed [17].

While technological advances in packaging continue to evolve, it is equally important to assess their environmental implications. Life Cycle Assessment (LCA), as standardised under ISO 14040/44 [18,19], offers a robust methodology to evaluate the potential environmental impacts of products or services throughout their life cycles. In the context of packaging systems, a comprehensive review by Bher and Auras [20], covering 113 LCA studies published between 2013 and 2023, revealed a strong research focus on rigid primary packaging (77%), compared to only 22% on flexible formats, predominantly plastics. Bottles and septic cartons were among the most frequently studied formats, while less than 10% of the studies addressed films. Moreover, most reviewed LCAs concentrated primarily on climate change and non-renewable energy use, often with divergent data quality, even for well-studied packaging formats. Notably, the inclusion of post-consumer recyclates in primary packaging was sparse [20]. Similar gaps were observed in the review by Bremenkamp and Gallagher [21], which focused on LCA studies of food packaging systems, particularly for ready-to-eat fish and meat products, from 2014 to 2023. Flexible packaging was underrepresented, and assessments involving recycled content were limited to PET trays. Of 41 comparisons across 13 LCA studies on plastics in the food sector, over 90% addressed drink bottles, cups, food service ware, and grocery bags [22], where increased recycled content consistently correlated with reduced climate change impacts and lower disability-adjusted life years.

Although limited in number, existing LCA studies on flexible multilayer packaging reflect similar trends. For example, structures such as MuMu (PET/metallised PET/PE) or MoMu (PP/metallised PP/PP) with higher levels of recycled content demonstrated environmental benefits across multiple impact categories [23]. Likewise, virgin films performed worse than films made from recyclates recovered via advanced solvent-based selective dissolution technology, when compared per kg of film. Benefits were observed for climate change, energy and water use, and toxicity impacts [24]. However, these two studies lacked detail on the intended applications of recyclates (e.g., food versus non-food). One study relied on simulated rather than empirical data, while another did not specify the source of waste used to produce recyclates and the recycling technologies employed.

Moreover, LCA studies on advanced recycling technologies for multilayer flexible packaging remain scarce and tend to focus on post-industrial waste, using processes such as deinking [25] or selective dissolution [26]. However, the applications of the recyclates in the next life cycle were not explored. Importantly, recycled low-density polyethylene (LDPE) and its linear variant (LLDPE) were only recently authorised for food-contact use under strict conditions, following Regulation (EU) 2022/1616, effective from October 2022 and amended in March 2025 [6]. The regulatory developments aforementioned, coupled with existing knowledge gaps, underscore the need for more targeted LCA studies that explore the environmental impacts of incorporating post-consumer recyclates, e.g., recycled PE, into flexible food packaging and evaluate the role of advanced recycling technologies in enabling such high-value applications.

This study responds to that need by addressing the environmental sustainability of integrating recycled LDPE (rPE), derived from household flexible food packaging waste, into new flexible food packaging applications. The work is aligned with recent regulatory frameworks and circular economy goals. Three interrelated sub-objectives were defined. First, the study focused on the material level by identifying the environmental hotspots (i.e., the processes or flows contributing the most to the environmental impacts) in the production of rPE pellets via advanced technologies, including tracer-based sorting (TBS), purification through delamination-deinking and deodorisation. The impacts of rPE pellets were then compared to those of virgin LDPE (vPE) pellets. Second, at the application level, the assessment focused on four multilayer (triplex) flexible packaging structures with equivalent functionality (serving as primary packaging for dry, high-barrier food products such as coffee powder). Two MoMu structures containing 34 and 50% rPE (by laminate surface weight) were compared against two virgin alternatives: one MuMu and one vPE-based MoMu. This aimed to determine whether shifting from MuMu to MoMu designs, and increasing the recycled content within MoMu, could lead to environmental benefits without compromising functional performance. Third, sensitivity, scenario and data quality analyses were conducted to explore how variations in key parameters influenced environmental outcomes, thereby ensuring the robustness of the results.

The remainder of this paper is structured as follows. Section 2 outlines the materials and methodology employed in the study. Key findings corresponding to the three aforementioned sub-objectives are presented in Section 3.1, Section 3.2 and Section 3.3. A broader discussion of the results, including study limitations, is provided in Section 4. Finally, Section 5 offers concluding remarks, policy-aligned recommendations, and suggestions for future research.

2. Materials and Methods

This LCA study was conducted following ISO 14040/44 standards through four steps: goal and scope definition, inventory analysis, impact assessment, and interpretation.

2.1. Goal and Scope Definition

This study focused on the innovative recycling route for PE-rich household flexible food packaging waste, along with design considerations for food packaging made from the resulting rPE pellets. The research was conducted within the framework of the Horizon Europe project “Circular Packaging for Direct Food Contact Applications” (Circular FoodPack, 2021–2024) [27]. The environmental impacts of the system were assessed at two levels: the pellet level and the packaging level, within a cradle-to-factory-gate system boundary. As shown in Figure 1, the pellet-level assessment covered all processes from waste collection to rPE pellet production. This allowed the identification of the environmental hotspots (i.e., specific processes or flows contributing the most to the environmental impacts), and hence provided a basis for improving the recycling route. At the packaging level, this study expanded the system boundary to include laminate conversion and packing processes, offering insights into the environmental sustainability of incorporating rPE pellets into flexible food packaging. The geographical scope of the assessment was limited to Europe.

2.1.1. Pellet-Level Analysis

rPE pellets were produced from household flexible packaging waste. As illustrated in Figure 1a, this PE-rich waste stream, comprising both food and non-food packaging with either MoMu or MuMu structures, was first separated from other waste fractions using near-infrared (NIR) sorting, followed by TBS. Tracers embedded during laminate conversion in food laminates during laminate conversion enabled the separation of food and non-food packaging. The food packaging fraction underwent a series of treatments: pretreatment, purification via delamination-deinking, recompounding and deodorisation. This tailored recycling route enabled the production of high-quality rPE pellets applicable to food packaging. The functional unit (FU) for the pellet-level assessment was defined as the supply of one kg of pellets at the factory gate.

2.1.2. Packaging-Level Analysis

The study progressed to evaluating the environmental impacts of flexible food packaging. Two triplex MoMu packaging types, primarily containing vPE purchased from the market and rPE produced from the aforementioned recycling route (see Figure 1a), were assessed: one with 50% rPE by laminate surface weight (FP1), representing the Circular FoodPack Consortium’s targeted design, and another with 34% (FP2), reflecting the reality of processability (particularly for sealing). They were compared to two virgin-based alternatives undergoing similar laminate conversion and packing processes and providing equivalent functionality: a MuMu composed mainly of PET, aluminium (Alu), and vPE (FP3), and a MoMu based solely on vPE (FP4), see Figure 1b. This comparison aimed to investigate whether a shift from MuMu to MoMu design, and additionally, increasing recycled content in flexible MoMu food packaging, could lead to environmental benefits without compromising functional performance. The FU for this packaging-level assessment was defined as the supply of one m² of packaging with the specified functionality described earlier.

Within the Circular Foodpack project, the four packaging structures were designed by experts from a global leader in the flexible packaging industry as part of work packages 6 and 7. All variants were developed to fulfil the same functional requirements, primarily food protection and food waste prevention. As part of work package 4, the design of FP1 and FP2 (with recycled content) was validated through approved packaging tests on their demonstrators, developed in work package 6, for damage resistance, sealability, and barrier properties, confirming the compliance with current food safety regulations [27]. Details of the studied packaging structures are presented in

Table 1. Overview of the four studied packaging structures.

Packaging	FP1	FP2	FP3	FP4
Structure	Triplex, MoMu	Triplex, MoMu	Triplex, MuMu	Triplex, MoMu
Thickness, µm	135.5	117.5	85.5	116.5
Surface weight g/m2	124.4	108.1	96.8	106.6
Outer (barrier) layer	OPV MDOPE-SiOx Ink & tracer Delamination primer	OPV MDOPE-SiOx Ink & tracer Delamination primer	PET Ink	MDOPE Ink
Middle layer	rPE	rPE	Alu	MDOPE-SiOx
Inner (sealant) layer	vPE & EVOH	vPE & EVOH	vPE	vPE
Adhesive	Solvent-free	Solvent-free	Solvent-based	Solvent-based
Recycled content	rPE: 50%	rPE: 34%	0%	0%
Virgin content	vPE: 42% (incl. MDOPE) Others: 8%	vPE: 57% (incl. MDOPE) Others: 9%	vPE: 56%, PET: 17%, Alu: 20% Others: 3%	vPE: 94% (incl. MDOPE) Others: 6%

.

2.2. Inventory Analysis

Inventory analysis involves compiling and quantifying the inputs (e.g., energy, materials) and outputs (e.g., products, co-products, waste, emissions to air, water, and soil) for the studied product system [18]. This includes directly related foreground processes and indirectly supporting background processes, as detailed below.

2.2.1. Foreground System

Primary mass and energy data for the foreground processes within the dashed frames in Figure 1 (i.e., the recycling pathway from waste collection to posttreatment, laminate conversion and packing) were provided by the Circular Foodpack industrial partners, and then validated via Mass Flow Analysis (MFA). Data gaps were filled using values from scientific literature and relevant databases. The involved technologies were developed at the Technology Readiness Levels (TRL) of 5–6, referring to technologies demonstrated (TRL 5) and validated (TRL 6) in an industrially relevant environment for key enabling technologies, as defined by the European Commission [28]. Details on TRL, mass and energy data, as well as process flow schemes (i.e., a step-by-step visual outline) for the foreground processes can be found in the Supplementary Materials, Tables S1–S3 and Figures S1–S3.

2.2.2. Background System

Secondary data were obtained from the ecoinvent v3.9 database, primarily the datasets for Europe (including infrastructure) when practically feasible, and supplemented with literature sources to model background processes, outside the dashed frames in Figure 1. These included the production of energy (electricity and heat), raw materials (e.g., virgin plastics: vPE, PET, EVOH, adhesives, chemicals, water), and waste treatment (e.g., incineration of solid residues with energy recovery, wastewater treatment).

2.2.3. Multifunctionality Modelling

Following the methodological framework of Sanabria Garcia et al. [29], the multifunctionality of plastic recycling, reflecting its dual role in both waste management and material production, was addressed using two complementary approaches. First, the substitution approach (also known as the credit or avoided burden method) was employed to deal with co-product multifunctionality. This approach considered the environmental net impacts (i.e., sum of burdens and credits) of avoiding the conventional end-of-life (EoL) treatment of PE-rich household flexible food packaging waste. In current European systems, this waste stream is typically collected with other household waste fractions through either mixed or selective systems. In the latter, it undergoes NIR and is sorted as a non-recyclable fraction, like in the mixed collection system, eventually directed to incineration with energy recovery (70%) and landfilling (details in Supplementary Materials, Section S1). Since household collection occurs similarly regardless waste is conventionally treated or recycled, this process was excluded from calculating the net impacts of the conventional EoL pathway to be avoided by recycling. Second, to handle the EoL multifunctionality of recycling (referring to the expanded lifetime of materials over multiple life cycles), the cut-off approach was applied. Under this approach, the burdens of material production are allocated to the previous life cycle, meaning the waste input enters the recycling system without associated upstream burdens (or is considered burden-free). Further methodological details on handling co-product and end-of-life multifunctionality are provided in Sanabria Garcia et al. [29].

2.2.4. Assumptions

The studied waste stream was assumed to be collected, sorted and recycled within France. Subsequently, the produced rPE pellets were assumed to be transported to Belgium for laminate conversion and then to Portugal for packing. Transport distances for waste collection and between processing sites were defined based on the potential future locations of these activities, as envisioned by Circular Foodpack Consortium partners in the relevant industry (see Supplementary Materials, Section S1).

All studied processes were tested at TRL 5–6; however, the mass and energy data were obtained under different conditions: some at the tested environmental, i.e., semi-industrial scale (TRL 6) for delamination-deinking, while others at full-industrial operation (e.g., pretreatment, laminate conversion, recompounding). Consequently, expert judgment from both industrial and academic members of the project consortium was used to estimate representative, industrially relevant data. For instance, solid residues generated during these processes were assumed to be incinerated with energy recovery, while wastewater from delamination-deinking was modelled as being treated via ultrafiltration, enabling closed-loop water reuse within this process. These assumptions aimed to ensure a consistent and comparable assessment between rPE and vPE pellets, as well as among the four studied food packaging structures.

2.3. Impact Assessment

The Environmental Footprint (EF) method, developed and recommended by the European Commission [30], was applied to assess the potential environmental impacts, performed in Excel and using Simapro software v9.5. Selecting key impact categories for analysis and in-depth discussions was guided by the draft Product Environmental Footprint Category Rules for flexible packaging [31]. They include climate change, fossil resource use, mineral-metal resource use, freshwater ecotoxicity, and particulate matter, which collectively account for approximately 80% of the environmental impacts associated with flexible packaging production. Acidification was also included due to its particular relevance to the recycling of PE-rich household packing waste, as highlighted in the literature [32,33].

2.4. Approach to Sensitivity, Scenario and Data Quality Analyses

2.4.1. Sensitivity Analysis Method

A parametric sensitivity analysis was conducted to identify the most influential parameters in the manufacturing of rPE pellets that drive their environmental performance. The scope of this analysis was limited to the pellet level for two reasons. First, pellets serve as a base material in the plastic industry; hence, focusing on this level offers opportunities for environmental improvement for both pellets and their multiple end-use applications beyond food packaging. Second, food packaging design is influenced by various factors, e.g., product type (dry vs. fresh), size, and commercial purpose (business or household) [34], that extend beyond varying layer thickness or recycled content alone.

Oracle Crystal Ball software v11.1 was used to perform Monte Carlo simulations. Parameters were varied over 10,000 iterations using a triangular distribution, each defined by ±10% of the original parameter values, following previous LCA studies [35,36]. An exception was made for process-efficiency parameters (defined as the mass yield, i.e., the ratio in mass of PE-rich output to input), where the upper value was limited to 100%, reflecting the maximum achievable process efficiency in practice (details in Supplementary Materials, Section S1).

2.4.2. Scenario Analysis Method

Based on the results of this sensitivity analysis, different scenarios were analysed to further evaluate their influence on the overall LCA results. Specifically, the effect of electricity sourcing was addressed through two scenarios: (i) European electricity dataset, and (ii) site-specific electricity datasets, depending on the locations where the studied processes were projected (details in Section 3.3).

2.4.3. Data Quality Assessment

The key process-based parameters identified as environmental hotspots at both pellet and packaging levels (Section 3.1 and Section 3.2) were evaluated for data quality. The Pedigree approach, widely used in the literature and adopted by ecoinvent and the International Reference Life Cycle Data System (ILCD, European Commission) [37], was applied, following the guidelines in the ecoinvent documentation [38]. For each parameter, five indicators were assessed: reliability, completeness, temporal correlation, geographical correlation, and further technological correlation. Each indicator was scored from 1 (highest certainty) to 5 (lowest certainty). These scores, combined with the default basic uncertainty factor of 1.05, were used to calculate the total uncertainty factor. Further details on the indicators, scoring criteria, and calculations are provided in the Supplementary Materials, Section S2.

3. Results

The environmental impacts regarding the six selected impact categories are presented for both recycled versus virgin PE pellets, as well as for MuMu packaging versus MoMu packaging with or without recycled content. These results are followed by the findings of the sensitivity, scenario, and data quality analyses.

3.1. Pellet-Level Hotspot Analysis

The net environmental impacts of rPE pellets arise from two main contributors (Figure 2). The first is the net impacts associated with the recycling pathway, including waste collection, sorting (NIR and TBS), and the recycling processes (from pretreatment, purification through delamination-deinking, to posttreatment with recompounding and deodorisation). The net impacts were evaluated as the sum of the environmental burdens from these processes, offset by the credits (or benefits) from energy (electricity and heat) recovered during the incineration of solid residues generated across the recycling pathway. Second, as outlined in Section 2.2, the model accounts for the avoidance of the conventional EoL treatment for PE-rich household flexible packaging waste, typically incineration with energy recovery and landfilling when this waste stream is recycled. Specifically, the avoided net impacts were calculated by summing the burdens of the conventional EoL pathway (excluding collection) and the associated recovered energy credits. This reflects the co-product multifunctionality of recycling, where the (net) impacts (i.e., sum of burdens and credits) of a displaced waste management pathway (only exclusive processes) are subtracted. Hereafter, this concept is referred to as “avoided conventional EoL” and is labelled as “avoided conventional EoL pathway (net, excl. collection)” in Figure 2 and Figure 3.

Climate change. Avoided conventional EoL offers substantial benefits by preventing CO2-eq emissions from avoided waste combustion, resulting in a net climate impact of -0.1 kg CO₂-eq/kg of rPE pellets, compared to 1.9 kg CO₂-eq for vPE. The negative value of the net impact does not imply that the studied recycling pathway acts as a carbon sink but reflects the climate benefits of accounting for the co-product multifunctionality of recycling in waste management and pellet production. However, purification introduces remarkable burdens in this category, largely due to its low process efficiency (80%, based on the mass ratio of PE-rich output to input). This inefficiency leads to a high mass of solid residues (or high material losses) sent to incineration. It is important to note that the burdens of incinerating these residues are included here, along with the credits from the energy they generate.

Fossil resource use. Avoided conventional EoL contributes to burdens rather than credits regarding this impact category. Additional burdens come from posttreatment (primarily recompounding), collection, and pretreatment. Nonetheless, the net fossil resource use for rPE (38.0 MJ per kg of pellets) remains 48% lower than that for vPE. Several factors explain this outcome. First, the waste feedstock consists primarily of PE films, which have a high calorific value. During its conventional EoL pathway, where 70% of the waste is incinerated, substantial energy can be recovered, effectively substituting fossil fuels in the EU grid energy generation. As a result, accounting for avoided conventional EoL can add burdens rather than credits in terms of fossil resource use. Second, high electricity demand in recompounding and pretreatment, along with the use of inefficient transport modes for waste collection (with diesel-fuelled lorries of 6-ton and 12-ton capacities, each representing 50% of the fleet)further contribute to the burdens. Note that electricity impacts were modelled using the EU electricity mix, predominantly fossil-based (hard coal), available in ecoinvent v3.9.

Mineral-metal resource use. All processes associated with the recycling pathway from collection to posttreatment remarkably contribute to the impact of rPE (12.2 mg Sb-eq/kg pellets versus 8.8 mg Sb-eq/kg for vPE). This is explained by similar factors as discussed under fossil resource use, including the electricity-intensive nature of pretreatment and recompounding processes, as well as the inefficient collection logistics relying on diesel-powered lorries. The current EU electricity mix is largely supplied by nuclear and natural gas, along with coal-based power, requiring high demand for these resources for infrastructure. Additionally, the use of natural gas-based heat in purification adds a further impact specific to this category. In contrast, the avoided conventional EoL, which plays the major role in climate change and fossil resource use, has a negligible influence on mineral-metal resource use. Keep in mind that this avoidance excluded waste collection, the process occurring in both pathways.

Freshwater ecotoxicity. This category follows a trend similar to that observed for mineral-metal resource use. All processes within the recycling pathway contribute notable burdens, particularly purification and collection, whereas the avoided conventional EoL negligibly contributes to the burdens, leading to substantially higher freshwater ecotoxicity for rPE (6.8 CTUe/kg pellets) compared to vPE (2.2 CTUe/kg pellets). These elevated impacts are primarily due to recurring factors: high material losses in purification, inefficient energy use during pretreatment and recompounding, and the reliance on suboptimal transport modes.

Particulate matter. The previously noted inefficient collection logistics contribute substantial burdens to particulate matter formation, which are further increased by burdens from the avoided conventional EoL. Nevertheless, rPE still performs environmentally better than vPE (4.4 × 10⁻⁸ and 6.9 × 10⁻⁸ disease incidences/kg pellets, respectively).

Acidification. A similar pattern to fossil resource use is noted for this category: the burdens stemming primarily from the avoided conventional EoL. In addition, all processes in the recycling pathway add burdens, resulting in similar acidification potential for rPE relative to vPE (7.2 × 10⁻³ mol ion H⁺ eq/kg pellets).

Overall, rPE pellets, produced from the advanced recycling with TBS, delamination-deinking purification and deodorisation, outperform vPE pellets regarding climate change, particulate matter formation, and fossil resource use, but worse regarding mineral-resource use and freshwater ecotoxicity. However, it is important to note that this comparison, based on the FU: one kg of pellets, does not account for potential quality degradation of rPE relative to vPE, nor its implications for downstream applications. Second, attention should be paid to reducing the material losses during delamination-deinking and energy use in pretreatment and recompounding, in addition to switching to lorries with higher capacity for waste collection. Which parameters influence the most the environmental performance of rPE pellets is further investigated in Section 3.3. Finally, the avoided conventional EoL (where incineration shares 70%) results in substantial environmental credits regarding climate change, but considerable burdens, as demonstrated in the cases of fossil resource use and acidification.

3.2. Packaging-Level Environmental Comparison

The packaging-level analysis explores the net environmental impacts of food packaging with MuMu design versus MoMu design with or without recycled content. Four packaging types with similar specified functionality (i.e., primary flexible packaging for dry food with high barrier requirements) were assessed, including three PE-rich MoMu structures: FP1 containing rPE at 50% of surface weight, FP2 containing rPE at 34%, and FP3 with vPE, as well as one MuMu structure (FP4) with only virgin materials. On the one hand, the environmental burdens of FP1 and FP2 primarily arise from three sources: (i) production of inputs attributed to collection, sorting, pretreatment, purification, posttreatment, laminate conversion and packing, (ii) incineration of solid residues generated across these processes, solvent oxidation and gas emissions during laminate conversion; and (iii) transport between sites. On the other hand, their environmental credits stem from energy recovery from residue incineration. As presented in Section 3.1, the avoided conventional EoL pathway (excl. collection) adds either burdens or credits for rPE present in FP1 and FP2.

The results, illustrated in Figure 3, show clear environmental advantages across all studied impact categories when shifting from virgin MuMu packaging (FP3), made from vPE, PET, and aluminium, to FP1 and FP2. In terms of climate change impacts, FP1 and FP2 achieve reductions of 65 and 67%, respectively, compared to FP3, which emits 1.05 kg CO₂-eq/m² packaging. FP1 performs slightly worse environmentally because its middle (rPE) layer had to be substantially thicker (68 µm compared to 40 µm in FP2) to meet the 50% recycled content target while the inner sealing layer must comply with strict safety and functionality standards required for food-contact applications, which limits how much its thickness can be reduced (from 50 µm in FP2 to 40 µm in FP1). This leads to higher rPE usage for FP1 and more material losses (residues) requiring incineration. Similarly, benefitting from the use of rPE, fossil resource use drops by 41 and 43% for both FP1 and FP2 compared to FP3, which demands 16.2 MJ/m² packaging. Regarding mineral-metal resource use, FP1 achieves a 16% reduction and FP2 a notable 28% reduction relative to FP3 (3.3 mg Sb-eq/m² packaging). Freshwater ecotoxicity impacts drop by 29% for FP1 and 35% for FP2, particulate matter formation decreases by 69% for FP1 and 72% for FP2 and acidification impacts reduce by 76% for FP1 and 78% for FP2, compared to FP3. The impacts on one m² of FP3 are equivalent to 5.0 CTUe, 5.1 × 10⁻⁸ disease incidences, and 6.8 × 10⁻³ mol ion H⁺, respectively.

Similarly, the vPE-based MoMu packaging (FP4) performs better environmentally than the MuMu FP3 across the six impact categories (i.e., a reduction by 50, 21, 17, 36, 65, 75%, respectively). This also implies that FP2 is remarkably less impactful relative to FP4 across the discussed impact categories, except for freshwater ecotoxicity where their impacts are similar. It can be explained by two reasons. First, FP2 and FP4 have similarities in their structures, including a triplex PE-based MoMu, thickness (117.5 and 116.5 µm, respectively), surface weight (108.1 and 106.6 µm/m², respectively), resulting in similar PE input (rPE and vPE: 91% in FP2 and 94% in FP4). Second, the considerably higher impacts of rPE against vPE per kg of pellets are observed for freshwater ecotoxicity (see Section 3.1 and Figure 2). Though vPE is less impactful than rPE also at mineral-metal resource use, it can be outweighed by a slightly higher PE input for FP4, resulting in its higher impacts against FP2. Similarly, FP1 performs better compared to FP4, except for mineral-metal resource use and freshwater ecotoxicity, where the impacts of FP1 are similar and higher by 11%, respectively. This finding is also linked to the aforementioned design choice: the need for a thicker rPE layer to achieve the targeted recycled content. Specifically, FP1 has a higher overall film thickness (135.5 µm) relative to FP2 (117.5 µm) and FP4 (116.5 µm), leading to greater total PE input, including both vPE and rPE. These findings highlight that, first, MoMu structures: FP1, FP2 and FP4 environmentally outperform FP3 in all considered impact categories. Second, incorporating rPE in MoMu flexible food packaging (FP1, FP2) provides additional environmental benefits compared to vPE-based MoMu (FP4) with similar structure and equivalent functionality, although there are some trade-offs for FP1 with high rPE content regarding freshwater ecotoxicity, where vPE performs substantially better than rPE.

Overall, to improve the environmental performance of flexible food packaging, recyclates should sufficiently substitute for virgin materials both in quantity and quality/functionality. This means decreasing the need for primary materials but also avoiding the ‘recycled content without functionality’. Recyclates are not simply added for the sake of appearance or meeting recycled content targets. Otherwise, thicker laminates become necessary, which leads to reduced environmental benefits on the one hand and increased processability challenges (e.g., in sealing) on the other hand. In food packaging, achieving high recycled content without increasing laminate thickness is especially challenging due to stringent food safety regulations, underscoring the importance of further research in this area.

Notably, laminate conversion is the dominant contributor to environmental impacts for both FP1 and FP2 across all six impact categories. FP1 and FP2 still rely on vPE (42 and 57% of their surface weight, respectively, see Table 1), preventing the packaging from achieving a fully closed-loop system. Additionally, residue incineration remains a significant source of climate change emissions, despite partial compensation from energy recovery. This underlines the need to improve the efficiency of the delamination-deinking process, as also noted in Section 3.1. Looking ahead, future efforts should focus on optimising laminate structures to allow for higher recycled content without compromising the functional performance and thereby avoiding thicker laminates. Also, shifting from virgin MuMu (FP3) to MoMu with (FP1, FP2) or without (FP4) recycled content while maintaining the functionality (e.g., serving as primary flexible packaging for dry food, requiring high barriers in this study), plays a key role in making flexible food packaging more environmentally sustainable and moving closer to a circular economy.

3.3. Findings from Sensitivity, Scenario and Data Quality Analyses

3.3.1. Sensitivity Analysis Results

The sensitivity analysis, conducted at the pellet level, examines how variations in parameters across the recycling chain (including waste collection and sorting) influence the environmental impacts of rPE pellets and the studied food packaging (Figure 4). It distinguishes between two key aspects: process efficiency and process impacts. Process-efficiency parameter refers to the mass yield (expressed as the ratio in mass of PE-rich output to input) of each process, considering only three processes: pretreatment (92%), purification (delamination-deinking, 80%), and recompounding (95%), as the others achieved 100% efficiency. For pretreatment and recompounding, when varying their efficiencies in a range of ±10%, the maximum value was higher than 1 and hence set at 1 (or 100% efficiency), see details in Supplementary Materials, Section S2. Process-impact parameters, on the other hand, represent the environmental net impacts of each process per kg of PE-rich input (e.g., kg CO₂-eq/kg of flakes to be purified for delamination-deinking). In Figure 4, since the contributions of the analysed parameters within each impact category sum to 100%, a higher contribution indicates a more influential parameter.

For climate change impacts, one of the most decisive factors is the avoided conventional EoL, which was considered to deal with the multifunctionality of recycling, as outlined in Section 2.2 and identified in Section 3.1 as the key factor substantially reducing the impacts of rPE pellets. Following closely is the pretreatment efficiency. Beyond that, purification efficiency emerges as the critical factor, resulting in a similar trend observed across the other five impact categories. Also, the recompounding efficiency proves influential to the impact variation. Changes in the pretreatment efficiency also contribute to variations in rPE impacts, particularly regarding fossil resource use and acidification. Additionally, variations in the process-specific impacts of delamination-deinking show a modest influence on mineral-metal resource use as well as freshwater ecotoxicity.

Overall, these findings highlight room for improvement. Key areas include enhancing the efficiency of delamination-deinking and recompounding, reducing electricity consumption, and shifting to renewable energy sources, which was further addressed via scenario analysis. It is important to note that delamination-deinking (with 80% efficiency) was tested at the semi-industrial scale. Although the study considered certain scale-up factors, such as incorporating ultrafiltration for wastewater treatment in the model, the measured process efficiency and energy consumption were not fully scaled. This underscores the importance of further testing these technologies under industrial conditions.

3.3.2. Scenario Analysis Results

Among the most influential parameters identified, variations in the efficiencies of pretreatment, purification (i.e., delamination-deinking), and recompounding are linked to multiple factors, e.g., technical feasibility, process scale, solvent types, hot washing temperature, etc. Variations in these factors might lead to changes in both mass and energy data, but more importantly, might be correlated, increasing uncertainty in modelling and forecasting. As a result, the scenario analysis explored the effects of different electricity sourcing options. Instead of using the electricity dataset for the EU (without Switzerland) available in ecoinvent v3.9, the scenario analysis considered specific sites of the involved processes for electricity production: France (FR) for waste management (collection, sorting and recycling), Belgium (BE) for laminate conversion, Germany (DE) for ink and tracer production, and Portugal (PT) for packing.

The results of this analysis are shown in Figure 5, where each marker represents the percentage change in impact for rPE pellets relative to vPE pellets and for MoMu (FP1, FP2 and FP4) compared to the reference: MuMu (FP3). The closer the markers are to each other along the Y-axis, the more similar their net impacts. In cases where markers overlap or are positioned very closely (e.g., mineral-metal resource use at pellet level, particulate matter and acidification at packaging level), it indicates that those options have nearly identical performance in that specific impact category.

At the pellet level, the site-specific scenario shows that rPE pellets can offer notably greater environmental benefits compared to vPE pellets, especially regarding climate change, acidification, and freshwater ecotoxicity. This improvement mainly stems from a change in electricity sourcing: from DE as the driver in the EU electricity mix, to FR in the site-specific case, where sorting and recycling take place. Electricity-related impacts per kWh are considerably higher in DE due to their reliance on hard coal and lignite, whereas the FR electricity mix is based mainly on oil, natural gas and nuclear sources. For the other three impact categories, where both DE and FR drive electricity-related impacts, no clear advantage emerges. The site-specific scenario reduces the benefits of rPE relative to vPE pellets regarding fossil resource use, mainly because FR heavy relies on nuclear energy, which has a higher burden in this category.

At the packaging level, however, the site-specific scenario does not lead to remarkable changes in the overall environmental impacts. This outcome can be explained by several factors. First, FP1 and FP2 still rely heavily on vPE, which lessens the influence of rPE-specific variations seen at the pellet level. Second, there is no shift in electricity sourcing for the production of virgin materials, such as vPE, PET, aluminium, adhesives, and inks, used in FP3 and FP4, meaning their environmental profiles remain unaffected by the scenario shift. Lastly, electricity use in laminate conversion, which takes place in BE, has a minor effect on the overall impact of packaging relative to the contribution of plastic-related flows.

3.3.3. Data Quality Findings

Key process-based parameters, including process efficiency (reflecting yield and material losses) and energy use across the following processes: pre-treatment, purification (via delamination and deinking), recompounding, and laminate conversion, were evaluated for data quality. These parameters and processes were identified as environmental hotspots in Section 3.1 and Section 3.2, with several also found to be among the most influential factors in the sensitivity analysis. The calculated total uncertainty factors, presented in Supplementary Materials, Section S2, indicate that the data for these key parameters are associated with low uncertainty.

4. Further Discussion and Study Limitations

Table 2. Summary of the key findings.

Pellet-Level Assessment:

Under the substitution approach for addressing multifunctionality of recycling, avoided conventional EoL pathway (70% incineration with energy recovery, 30% landfill, excl. waste collection) provides climate benefits but adds burdens in fossil resource use, particulate matter, and acidification, Recycled PE pellets outperform virgin PE in climate change, fossil resource use, and particulate matter.

Packaging-level assessment:

Switching from virgin MuMu to MoMu with/without recycled content improves environmental performance across the six impact categories, Higher recycled content (e.g., 50%) may require thicker laminates, increasing environmental impacts and processing challenges.

Rooms for improvement:

Reduce material losses in delamination–deinking, Lower energy use in pretreatment and recompounding, Enhance waste collection logistics.

summarises the key findings from Section 3.1, Section 3.2 and Section 3.3, followed by discussions on methodology (Section 4.1), regulatory-aligned packaging design (Section 4.2), and study limitations (Section 4.3).

4.1. Methodological Aspects

Choosing an appropriate FU in comparative LCA is crucial for generating reliable and meaningful results. For plastic products, rather than relying on mass alone, the FU should reflect the product’s actual functionality in its end-use application, or the key material properties relevant to that application [39]. This study focuses on plastic recyclates at both pellet and packaging scopes. For the former, the FU of one kg of pellets at the factory gate (Section 3.1) does not reflect the difference in quality (or functionality) between rPE and vPE pellets; hence, it aims more at identifying the hotspots and rooms for improvement across the recycling chain rather than comparing the environmental impacts of rPE and vPE pellets. Focusing on food packaging, the FU of one m² of packaging with specified functionality gives a more complete picture when considering the recyclates’ application specifications (i.e., primary flexible packaging for dry, high-barrier food). Also, the impacts were estimated for packaging rather than just those of laminates, including energy use and material losses (5%) during the packing step.

Beyond the FU definition, methodological choices, e.g., system boundaries, inventory data, impact assessment methods, and especially allocation approaches, can strongly influence LCA outcomes, as shown across various sectors, including packaging [40] and plastic recycling [29,41]. Among these, multifunctionality modelling remains one of the most debated topics in LCA research [42]. In the context of plastic recycling, Sanabria et al. [29] demonstrated that choices for handling co-product and EoL multifunctionality can substantially affect results. For instance, results framed from a product perspective (e.g., FU: supply of one kg of recycled pellets, as in this study) are not directly comparable to those from a waste perspective (e.g., FU: treatment of one ton of waste). However, results can be converted using the recycling yield. Importantly, comparing absolute results across studies with different methodological choices is discouraged, as it may lead to misinterpretation [29]. This study applied substitution and cut-off approaches to address co-product and EoL multifunctionality (see Section 2.2). Crucially, when applying substitution to account for avoided impacts of an EoL strategy, e.g., recycling versus conventional disposal in this study, it is crucial to consider the entire waste management pathway, from collection to treatment. Focusing on treatment alone can lead to double-counting of shared processes like collection, which has not been thoroughly addressed in the literature.

Moreover, the choice of alternative to be avoided by the studied EoL strategy, whether conventional disposal (e.g., 70% incineration and 30% landfilling), incineration or landfill alone, can influence the results. For instance, using incineration alone has a minimal effect at the pellet (rPE) level, whereas assuming only landfilling leads to a greater impact. This is especially evident in categories driven by the incineration of high-calorific, PE-rich waste, such as climate change, fossil resource use, and acidification (Section 3.1). At the packaging level, the effect is less pronounced due to the dominant contribution of virgin materials, particularly vPE. Further details are provided in Supplementary Materials Section S3. However, it should be noted that the discussed impact categories do not capture key environmental risks of PE waste landfilling. In the context of the EU Waste Framework Directive [43], landfilling is considered the least preferable option in the waste hierarchy due to long-term environmental concerns.

4.2. Food Packaging Design in Alignment with the European Regulatory Framework

In the context of flexible packaging, material thickness design is closely linked to multiple application-specific factors, including material selection, mechanical properties (e.g., tensile strength, stiffness), barrier properties (e.g., oxygen and moisture permeability), machine compatibility, economic cost, consumer appeal [44], and EoL recyclability [45]. For the packaging structures studied, thickness design was primarily driven by their intended application (serving as primary packaging for dry, high-barrier food products such as coffee powder) and the associated functional requirements, particularly mechanical strength and barrier performance. In addition, the incorporation of rPE into FP1 and FP2 further influenced thickness design, especially in terms of machine compatibility (e.g., sealability) and regulatory compliance with food safety standards. The Food and Drug Administration (FDA) recommends that a virgin material layer of sufficient thickness can effectively prevent contaminant migration from non-food-contact recycled layers, based on authorised migration tests [46]. To achieve the 50% rPE content target while maintaining food safety and functional performance, FP1 requires a thicker middle (rPE) layer. As a result, its inner (sealant) layer, composed solely of virgin materials (vPE and EVOH) and in direct contact with food, also needs to be sufficiently thickened. Both FP1 and FP2 have similar thicknesses in their outer virgin MDOPE layer, ensuring the required barrier performance. Consequently, FP1 needs more material input, which leads to worse environmental performance relative to FP2, with 34% rPE content, across all six impact categories. Sealing was also more challenging for FP1 due to its higher total thickness.

Additionally, according to Regulation (EU) 2025/40, packaging is classified into four groups, based on material composition and intended use [4]. The food packaging types studied here fall under the third category: contact-sensitive packaging made of non-PET plastics as the major component. FP1 and FP2, with a recycled content of 50% and 34%, respectively, meet the EU targets set for this packaging category: 10% recycled content by 2030 and 25% by 2040. However, it is important to note that these targets were set for contact-sensitive plastic packaging in general, without differentiating between food or non-food applications. Interestingly, the findings in this work do not align with those of Tunçok-Çeşme et al. [23], who reported that higher recycled content generally correlates with lower environmental impacts across numerous impact categories. The reason for this is that their analysis did not consider recyclate quality and application-specific requirements. This study supports the fact that simply increasing recycled content is not a silver bullet [47]. To mitigate, e.g., climate change impacts, it is essential to factor in recyclate quality and recycling efficiency. Also, substitution rates of recyclates for virgin materials, and consequently LCA results, can vary due to application-specific requirements and uncertain market dynamics [41]. This study captures the effect of the former but does not address the latter. Crucially, recyclate quality and intended application matter greatly when comparing recycling technologies. For example, high-quality pellets from advanced mechanical recycling with deinking might only replace 20% of virgin materials in high-value applications, while lower-quality pellets from conventional technology can substitute up to 80% in low-value ones [25].

Mechanical recycling currently plays a vital role in extending the life cycle of plastics and reducing reliance on virgin fossil-based materials; however, it often degrades material quality and properties, limiting the suitability of recyclates for high-quality applications [48,49]. Within the context of a circular economy, this raises a critical question about how waste generation can be minimised while maintaining the value (or functionality) of materials in the economy for as long as possible. Efforts to address this challenge are supported by recent EU regulations, such as Regulation (EU) 2024/1781 and 2025/40 of the European Parliament and the Council. Regulation 2025/40 sets targets to reduce packaging waste per capita by 5% by 2030 and 15% by 2040 (relative to 2018 levels) [4]. Regulation 2024/1781 introduces comprehensive ecodesign requirements aimed at enhancing product durability, reparability, recyclability, reusability, and material efficiency across the EU market [50]. Beyond the regulatory support, a range of strategies has been proposed in the literature to further improve the quality of plastic recyclates. They include optimising product design to minimise material heterogeneity and deploying advanced sorting and recycling technologies [51,52]. Moreover, innovative approaches are emerging, such as precise quantification of recycled content in plastics and the adoption of chemical recycling methods for polymers with complex compositions. The application of modifiers, e.g., compatibilisers, coupling agents, during plastic re-extrusion has also been shown to substantially improve the mechanical and functional properties of recyclates. These advancements are comprehensively reviewed in Soomro et al. [53] and Ghosh [54].

In this context, two key design approaches have gained prominence: design-for-recycling and design-from-recycling. The former is well established in both scientific and industrial domains with a focus on enhancing product recyclability at the EoL stage via improving factors such as recovery value and technical feasibility for recovering products and their components [51,52]. In contrast, design-from-recycling assesses the feasibility of manufacturing new products from recyclates and is currently underrepresented [51,55]. This study highlights the interconnection between these two design strategies. Specifically, during design-from-recycling for the studied food packaging with rPE (FP1 and FP2), key design-for-recycling principles were also integrated to support closing the material loop. Specifically, MoMu structures, known for better recyclability [10,12], were adopted. Additional design measures were implemented, including tracers to improve sorting efficiency, delamination primer and solvent-free adhesives to enable the water-based delamination-deinking process using cost-effective NaCl-based detergents. These measures enhance rPE quality, allowing partial replacement of vPE in new laminates without compromising functional performance. Importantly, packaging produced in this way can be efficiently recycled again using the same technologies, reinforcing the potential for a closed-loop system in flexible food packaging.

As the packaging industry increasingly shifts from MuMu to MoMu structures, our findings reinforce the environmental value of this transition: MoMu packaging, whether composed entirely of vPE (FP4) or blending vPE with rPE (FP1, FP2), shows clear environmental benefits (Figure 5). However, as aforementioned, the high recycled content (50%) can lead to the need for thicker laminates, which increases challenges in processing and decreases the environmental benefits compared to the lower one (34%). Importantly, this study underscores the infeasibility of completely closing the loop for food packaging at present due to not only material losses during the recycling chain but also challenges in design and processing to further increase recycled content. However, the study also underscores that, despite advanced sorting and recycling technologies (i.e., TBS and purification via delamination-deinking), the inherent quality loss in mechanically recycled PE remains unavoidable. This limits full circularity, resulting in a partially closed-loop recycling system instead. Also, mechanical PET recycling is an established and authorised process, whereas mechanically recycled plastics other than PET do not yet have recycling processes approved by the European Food Safety Authority (EFSA) for food contact use, addressed in Regulations (EU) 2022/1616 [6] and 2025/351 [5]. In that context, chemical recycling, which enables the recycling of mixed post-consumer plastic waste to high-quality recyclates suitable for contact-sensitive applications [56,57], could be part of the solution to further improve the environmental performance of food packaging. Consequently, further research on integrating chemically recycled plastics in food packaging could be crucial.

Lastly, beyond technological innovation, environmental taxation plays a crucial role in advancing environmental goals [58]. Defined in Regulation (EU) 691/2011 and the European System of Accounts (ESA), environmental taxes are levied on physical units with proven negative environmental impacts [59]. In the context of municipal waste management, taxes on landfilling and incineration have been shown to reduce environmental burdens, e.g., climate change, particulate matter, and ecotoxicity, as observed in Spain [60]. At the EU level, such taxes have effectively shifted waste away from landfill toward incineration and recycling [61]. While their short-term impact on recycling rates and circularity appears mixed, long-term effects are expected to be positive [62]. Furthermore, a cost–benefit analysis of packaging waste management in Portugal, Belgium, and Italy revealed that the operational costs of selective collection and sorting are often offset by the avoided costs of alternative EoL treatments [63]. To enhance these outcomes, incentive mechanisms should be developed to support environmentally friendly technologies and processes [58]. In line with this, Regulation (EU) 2024/1781 on ecodesign for sustainable products recommends tools such as eco-vouchers and green taxation to encourage sustainable consumer behaviour [59].

4.3. Study Limitations

This study gives insights into the use of recyclates in food packaging applications; however, several limitations exist. First, the assessment was limited to a cradle-to-gate scope. Although FP1 and FP2 were designed for recycling, their EoL recyclability and related environmental benefits were not evaluated. While MoMu is theoretically more favourable than MuMu due to the non-recyclable nature of MuMu, this does not guarantee greater environmental sustainability, especially given the gap between technical recyclability and actual recycling rates for plastic films [21]. Second, the advanced technologies employed (e.g., TBS, delamination-deinking, and recompounding) were assessed at the TRL of 5–6. Although their primary mass and energy data were collected and scaled (whenever practically feasible) to full-industrial operation for this assessment (details in Section 2.2), further testing at higher TRLs is essential to validate these findings. Third, the study focused on functional equivalence and environmental impacts; hence, future work should include economic and social (e.g., consumer preference) aspects in optimising packaging designs towards sustainability [64].

5. Conclusions

The environmental performance of recycled PE pellets, produced from food-grade fraction sorted from PE-rich household flexible packaging waste, using tracer-based sorting, and delamination-deinking, shows clear advantages over virgin PE pellets regarding climate change, fossil resource use and particulate matter. This climate benefit is primarily driven by substantial credits from the avoided conventional end-of-life (EoL) pathway, comprising 70% incineration with energy recovery and 30% landfilling. However, this avoidance adds burdens rather than credits regarding fossil resource use and particulate matter. This is largely due to the incineration of high-calorific PE-rich waste, which emits considerable greenhouse gases and air pollutants but also offsets fossil-based grid energy in Europe, leading to trade-offs. A similar pattern is observed for acidification, where avoiding the conventional EoL pathway also increases burdens. It is important to note that when addressing multifunctionality through the substitution approach, shared processes, such as waste collection, between the recycling and displaced EoL pathways must be excluded to avoid double-counting.

In contrast, for freshwater ecotoxicity and mineral-metal resource use, where the avoided conventional EoL pathway contributes minimally, virgin PE pellets outperform recycled PE ones. This is primarily due to inefficiencies across the recycling pathway, including the 80% yield in delamination-deinking, which sends a substantial amount of solid residues to incineration, and energy intensity in recompounding and pretreatment, along with suboptimal collection logistics. Recycled PE pellets also show similar acidification impacts to virgin ones, reflecting high contributions of the avoided conventional EoL pathway and the operational inefficiencies aforementioned. The functional unit used for this comparison (i.e., one kg of pellets), however, does not reflect the differences in pellet quality or the suitability for downstream applications. Environmental impacts of this recycling pathway can be enhanced through reducing the material losses in delamination-deinking, lowering energy use in recompounding and pretreatment, as well as switching to more efficient logistics for waste collection.

When applied to flexible food packaging, shifting from virgin MuMu to MoMu structures with or without recyclates, while preserving the functional performance (specifically, for dry, high-barrier food products like coffee powder), gains environmental benefits across the six key impact categories to different extents. However, recyclates should adequately replace virgin materials in both quantity and quality. Otherwise, higher recycled content (e.g., 50% versus 34%) may require thicker laminates; this increases challenges in processing (e.g., sealing) and the environmental impacts of packaging, particularly for freshwater ecotoxicity, and mineral-metal resource use. This design choice also substantially influences climate change; however, the effect reduces when accounting for the avoided conventional EoL pathway, which offers credits to partially offset the trade-offs.

These findings also align with recent EU regulatory frameworks, such as Regulation (EU) 2025/40 and 2024/1781, which promote packaging waste reduction and ecodesign. This study highlights the need to integrate both design-for-recycling and design-from-recycling perspectives in laminate development, making it relevant for multiple stakeholders (waste managers, recyclers, laminate producers, food manufacturers, and policymakers). Achieving a fully closed-loop system for food packaging remains unfeasible due to high material losses during recycling and challenges in design to further increase mechanically recycled content. Even with advanced sorting and purification (e.g., delamination-deinking), recyclate quality degradation persists, resulting in a partially closed loop. Future research, therefore, should prioritise assessing sensitive-contact packaging (including both food and non-food) with varying recycled content, and benchmarking advanced technologies, especially chemical recycling, which delivers high-quality recyclates applicable for such applications.

Industrial-scale testing is crucial to optimise process efficiencies and minimise environmental burdens. Further experimental research into process parameters, such as deinking agent concentration, solid/liquid ratios, and operational conditions, can enhance recycling performance. Advancing these recycling technologies to higher Technology Readiness Levels (TRL 8-9) will be essential to validate their technical robustness and suitability for industrial-scale applications. Complementary economic instruments, e.g., environmental taxation, avoided waste management costs, and eco-incentives proposed under recent regulations, can further accelerate the shift toward sustainable food packaging and circular economy goals.