From Crude to Green: The Environmental Benefits of Bio-Oil in Flexible Polyurethane Foams

Abstract

Flexible polyurethane foam (PUF) is a vital material across diverse applications, and its global market is projected to continue growing. Driven by regulatory and consumer demand for sustainable materials, the PUF industry is exploring alternatives to petroleum-derived raw materials, such as vegetable oil-derived bio-polyols. Although bio-based alternatives to fossil-derived foams have been developed, their environmental benefits remain to be fully assessed. Therefore, this study evaluates the environmental performance of flexible PUF production by comparing a conventional fossil-based formulation with a bio-based alternative using a cradle-to-gate Life Cycle Assessment (LCA). The bio-based PUF reduced global warming (6%), fossil resource scarcity (9%), and mineral resource scarcity (6%), but caused significant increases in freshwater eutrophication (91%) and marine eutrophication (19%), mainly due to agricultural processes associated with soybean cultivation. Regardless of the formulation, polyol and toluene diisocyanate production were identified as major environmental hotspots. These results highlight both the decarbonization potential and the trade-offs of bio-based raw materials, underlining the complexity of achieving sustainable PUF production. Overall, the findings provide quantitative insights to guide more sustainable design and sourcing strategies for flexible PUF in the transition from fossil to renewable feedstocks.

1. Introduction

The global flexible polyurethane foam (PUF) market has experienced significant growth due to its wide-ranging applications across industries. In 2023, this global market was valued at USD 42.7 billion and is projected to reach USD 56.9 billion by 2028, at a compound annual growth rate (CAGR) of 5.9%. This expansion is primarily driven by the increasing demand in the automotive sector and the furniture and bedding industry, where flexible PUF provides key advantages in comfort and durability [1].

Regarding the Europe, Middle East and Africa (EMEA) region, in 2023, over 1.98 million tonnes of flexible PUF were produced, and projections indicate a CAGR of 3.5% by 2028, despite challenges like the fluctuation of raw material prices and supply chain disruptions. Notably, the bedding and furniture sectors accounted for 84.2% of total flexible polyurethane (PU) consumption in EMEA [2], highlighting the significant impact of these industries on material usage and respective end-of-life (EoL) [3].

The projected expansion of the PUF market is expected to increase the dependence on fossil-based raw materials and associated emissions. Consequently, exploring sustainable alternatives becomes essential to mitigate such environmental pressures. Therefore, the interest in sustainable alternatives, such as bio-based and CO₂-derived polyols, has also been consistently growing. Even though the industrial implementation of alternatives to fossil raw materials remains limited due to high production costs and technical constraints, the flexible PUF industry is looking for greener options because of the increasing consumer demand for safer and environmentally friendly products and stricter regulatory frameworks. Of major relevance is the European Chemicals Agency (ECHA), which has introduced specific restrictions on the use of diisocyanates under the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH) regulation via the Commission Regulation (EU) 2020/1149 of 3 August 2020, which imposes stricter controls on diisocyanates due to their adverse health effects [4].

In this context, the flexible PUF industry is increasingly exploring renewable raw materials to reduce dependence on fossil resources and lower carbon emissions [1]. Despite all the efforts that have been made, renewable polyols only replace partially conventional counterparts in industrial formulations [5]. Furthermore, systematic life cycle assessment (LCA) studies regarding the benefits of using renewable polyols are scarce and generally limited to laboratory or pilot scale. At the same time, alternatives such as non-isocyanate polyurethane foams (NIPUF) remain technologically challenging and not yet viable at large scale. These emerging systems face difficulties in achieving comparable mechanical properties, curing kinetics, and industrial processability. Nevertheless, recent advances in catalytic routes and the use of cyclic carbonates and amines look promising and may overcome these limitations in the near future though LCA studies still need to be properly carried out [6].

LCA is an internationally standardised methodology for assessing environmental impacts throughout a product’s life cycle [7,8]. This methodology has been used to assess the potential environmental impacts of PUF-derived products [9]. An Environmental Product Declaration (EPD) has been developed to assess the environmental performance of fossil-based flexible PUF, based on a cradle-to-gate perspective [10]. Lanoë et al. [11] evaluated the environmental impact of mattresses produced from flexible PUF block based on a cradle-to-grave perspective, including the comparison between three EoL treatment options: landfill, recycling, and incineration. These studies suggest that the impacts of PUF production vary significantly depending on raw material sources, formulation choices, and production technologies. Moreover, while several LCA studies have assessed PUF-related products, comprehensive evaluations of the environmental performance of flexible PUF—particularly those produced using bio-based feedstocks—remain scarce. To the best of our knowledge, no published LCA study has comprehensively assessed the environmental performance of industrial-scale flexible PUF produced from bio-oil under real manufacturing conditions. Industrial-scale data reflect representative real-world production processes, thereby enabling a more accurate and reliable assessment of the potential environmental impacts of a product compared to data derived from development scales which are not fully optimized. Therefore, this study aims to address this gap by performing a cradle-to-gate LCA of industrial fossil-based and bio-based flexible PUF produced in Portugal. The study explores the potential benefits of increasing the share of bio-based polyols, while preserving key technical properties such as density, hardness, elasticity, tensile strength, and air permeability. This study addresses the following research questions: (1) Does bio-based flexible PUF reduce climate-related impacts compared to conventional flexible PUF? (2) What environmental trade-offs emerge from substituting fossil-based polyols with bio-based alternatives? (3) Which life cycle stages contribute most to the overall environmental impacts observed?

2. Materials and Methods

The LCA methodology was applied following the ISO 14040 and 14044 standards [7,8].

2.1. Goal and Scope Definition

The goal of the present study is outlined in the Introduction section, and the main elements of the scope, including those related to the functional unit (FU), system boundaries and multifunctionality, are addressed below.

The production of flexible PUF for mattresses and upholstery primarily relies on slabstock PUF, tailored to meet diverse market demands through various specifications and formulations, typically varying densities between 16 to 80 kg/m³, striking a balance between comfort and support for everyday use [12].

Flexible PUF foam production typically involves: (a) the mixing and reaction of isocyanate, polyol, and water and (b) block formation. Flexible foam formulation is strongly influenced by perceptions of comfort, which are related to the foam’s stiffness, resilience, and breathability (air permeability) [13].

Two flexible PUF industrially produced (

Table 1. Flexible PUF formulations considered.

Raw Materials	Unit	PUF Formulation
		Conventional	Bio-Based
Polyol mix	g	645.0	584.0
Bio-polyol	g	-	103.0
TDI—isocyanate	g	321.0	271.0
Water—blowing agent	g	26.0	21.9
Silicones—surfactants	g	6.4	7.5
Tertiary amines and tin salt—catalysts	g	1.8	2.1
Pigments	g	0.20	10.3

), produced by a flexible PUF representative company in Portugal, were considered:

(1)

Conventional flexible PUF—using 100% fossil-polyol mix;

(2)

Bio-based flexible PUF—incorporating 15% bio-polyol and 85% fossil-polyol mix.

The fossil-polyol mix (Table 1) consists of several polyols, mainly long-chain polyether polyol (a combination of polyethylene oxide polymer, propylene oxide, and glycerol), reinforced styrene-acrylonitrile (SAN) polyether polyol (a mixture of styrene-acrylonitrile polymer and polyethylene oxide, propylene oxide, and glycerol) and polypropylene glycol. Variations in formulation, such as types of polyols, isocyanate, and water, can result in differences in the physical properties of the flexible PUF foams. Both formulations were designed and industrially optimized to ensure equivalent mechanical and comfort performance. Although detailed physicochemical data for the raw materials are confidential, all polyols employed fall within the typical hydroxyl number range for flexible PUF applications, ensuring comparable reactivity and foaming characteristics. The bio-based polyols used present hydroxyl values and viscosities consistent with those of the conventional petrochemical counterparts, thereby maintaining functional equivalence. Minor adjustments to catalyst and surfactant concentrations were implemented to fine-tune the mechanical properties and processing behaviour. Consequently, any observed differences between the foam systems can be attributed to the feedstock origin rather than to discrepancies in final product performance.

For the present study, the physical properties of the PUF prepared using the two formulations are summarized in

Table 2. Physical properties of the flexible PUF under study.

Property	Method	Unit	Flexible PUF
			Conventional	Bio-Based
Density	ISO 845:2006 [14]	kg/m3	22.8–26.0	29.3–31.8
Resilience	ASTM D3574-17 [15]	%	42.3–49.9	43.3–55.0
IFD 25%	ISO 2439:2008 [16]	N	118.0–156.0	93.0–112.0
IFD 40%	ISO 2439:2008 [16]	N	150.0–192.0	117.0–140.0
IFD 65%	ISO 2439:2008 [16]	N	277.0–369.0	215.0–269.0
Sag factor	-	-	2.35–2.37	2.31–2.40
Compression hardness [strain 40%]	ISO 3386:1986 [17]	kPa	3.5–4.7	2.8–3.4
Elongation at break	ISO 1798:2008 [18]	%	112.0–263.0	180.0–209.0
Tensile strength	ISO 1798:2008 [18]	kPa	107.0–166.0	114.0–118.0
Air permeability	ISO 9237:1995 [19]	L/m2/min	636.0–1320.0	666.0–1030.0
Fatigue—hardness loss	ISO 3385:1989 [20]	%	24.4–32.0	26.6–27.7
Fatigue—thickness loss	ISO 3385:1989 [20]	%	2.0–4.0	2.2–3.4

.

Density directly affects cost and load-bearing capacity and influences foam’s mechanical behaviour and performance, perceived by consumers as comfort. The bio-based flexible PUF formulation, which has a lower content of blowing agent (water) and Toluene diisocyanate (TDI), generates less in situ carbon dioxide (CO₂) during the reaction between water and isocyanate. This results in a higher apparent density and reduced stiffness (Table 2) compared to the conventional foam, primarily due to the lower formation of urea linkages during reaction (Figure 1) [13]. Despite these differences in composition and properties, both formulations maintain the essential performance characteristics required to be classified in the same functional category as flexible PUF.

Specifically, the resilience, tensile strength, and fatigue resistance (regarding both hardness and thickness loss), confirm functional equivalence. The ranges reported for each property reflect the variability inherent to industrial slabstock foam production. This dispersion arises from batch-to-batch fluctuations in raw materials and minor, unavoidable variations in process parameters such as conveyor speed, mixing temperature, and curing conditions. In particular, the wide range observed for air permeability (636.0–1320.0 L/m²/min for the conventional foam and 666.0–1030.0 L/m²/min for the bio-based foam) is typical of large-scale continuous processes, where foam structure and cell openness may slightly vary along the production line [13]. Such variability is therefore not anomalous but representative of real industrial manufacturing conditions, confirming the robustness and representativeness of the data used in this study.

Given that the foam’s physicomechanical properties, such as stiffness, resilience, and breathability, directly influence functional performance, it is essential to assess how these characteristics affect the comfort perceived by the end-users. This remains a key evaluation criterion in flexible foam applications.

Foam stiffness is commonly evaluated through indentation force deflection (IFD) and compression hardness. IFD measures the force required to indent the foam to a specific percentage of its thickness, providing insights into its firmness and load-bearing properties. Standard IFD measurements are taken at 25%, 40%, and 65% indentation levels, allowing for a comprehensive evaluation of the foam’s behaviour under different compression stages. A key indicator of comfort and support is the support factor, also known as the sag factor. This factor is calculated as the ratio of the IFD at 65% indentation to the IFD at 25% indentation, reflecting both initial softness and long-term support. A sag factor between 2.0 and 2.5 is typically associated with foams that balance comfort and durability, making them ideal for seating and bedding applications [21]. ISO 2439 provides a standardized method for measuring IFD [16].

The resilience measures the foam’s rebound elasticity, which is crucial for maintaining dynamic support and responsiveness. The foam’s rebound elasticity is measured following the standard ASTM D3574, Test F, and is also known as ball rebound [15].

Breathability, defined as the foam’s ability to allow air to pass through its structure, is a key factor in enhancing comfort by promoting ventilation and reducing heat buildup. Air permeability tests quantify this property, which evaluates the foam’s porosity by measuring airflow resistance within its cellular structure. The standardized method for assessing air permeability in flexible PUF is outlined in ISO 9237 [19].

Compression hardness assesses: (i) stress–strain behaviour under compression, ensuring adequate support and preventing sinking, and (ii) tensile strength and elongation to ensure durability under stress, assessing elasticity and tensile capacity, and (iii) dynamic fatigue testing evaluates foam durability and property retention over time, vital for long-term performance and user satisfaction.

Given that comfort-related parameters are key to the functional performance of flexible PUF in end-use applications such as furniture and bedding, the FU defined in this study is the production of 1 kg of flexible PUF at the industrial scale. This reference was chosen to ensure consistency with available inventory data and comparability with previous LCA studies on flexible PUF. Although the conventional and bio-based formulations exhibit slightly different densities, both systems were designed and optimized to achieve equivalent mechanical and comfort performance (e.g., sag factor, resilience, and fatigue resistance). These adjustments satisfy the functional equivalence requirement defined in ISO 14044, validating the use of a mass-based FU as a reliable and comparable basis. This approach also allows for a consistent comparison of environmental impacts while maintaining relevance to real-world usage scenarios. An LCA approach from cradle-to-gate, i.e., from the production of raw materials to flexible PUF, was applied. Figure 2 shows the system boundaries to produce flexible PUF. The system boundaries encompass the extraction and processing of raw materials, and the manufacturing stage consists of mixing and reaction, the flexible PUF formation (designated as “slabstock” or simply “block formation”), and the treatment of solid waste generated from manufacture. The construction and maintenance of capital goods in the form of buildings, machinery, and equipment are outside the scope of the system.

The production of flexible slabstock PUF typically follows a semi-continuous process that minimizes manual handling and ensures uniform product quality. In this industrial configuration, the formulation components, polyols, isocyanates, catalysts, surfactants, blowing agents, pigments and additives, are continuously metered into a mixing head, where they are blended and immediately poured onto a moving conveyor. As the reaction proceeds, the foam expands and cures while advancing along the conveyor line, which is enclosed within a ventilated chamber designed to capture process gases and maintain safe operating conditions. Although machinery design may differ among manufacturers, the overall production principle remains consistent across facilities. This approach, widely implemented in Europe, represents standard practice in modern flexible PUF manufacturing [10]. Figure 3 illustrates the main stages of this process.

Although this study focuses on the production of flexible PUF, certain upstream processes in its supply chain are multifunctional. For example, soybean oil extraction generates both the oil used in polyol production and by-products such as soybean meal. Similarly, petroleum refinery produces naphtha alongside other co-products, which is used to produce polyol [22]. The data for these processes were obtained from the Ecoinvent v3.9.1 database [23], and the allocation of input and output inventory data among by-products was determined using the economic allocation factors applied in Ecoinvent. This approach represents the market-driven relationships among co-products in the value chain. In processes such as soybean crushing or crude oil refining, product flows are primarily governed by economic value rather than physical mass or energy content. Therefore, economic allocation provides a realistic reflection of these industrial systems and aligns with established LCA practice for market-oriented processes.

2.2. System Description

The bio-polyols used in this study are commercially available products derived from refined vegetable oils, primarily soybean oil. According to the manufacturers’ technical documentation, they are obtained through functionalisation processes, although specific synthesis details are not publicly available. To model the upstream renewable feedstock, the dataset “vegetable oil, refined soybean oil, to generic market” from Ecoinvent v3.9.1 was used, representing the cradle-to-gate production of refined soybean oil. This approximation provides a consistent and transparent representation of the renewable feedstock stage within the system boundary and is appropriate given the study’s focus on comparing industrially relevant commercial formulations.

As mentioned in Section 2.1, the manufacturing process of flexible PUF involves a mixing and reaction stage followed by block formation. Raw materials such as polyols and TDI are stored in tanks, while surfactants, catalysts, and pigments are kept in metallic drums and intermediate bulk containers (IBC), the latter consisting of a High-Density Polyethylene (HDPE) tank enclosed in a steel cage. Both the IBC and metallic drums are supported on wooden pallets. Nitrogen is used to pressurize raw material tanks to maintain proper flow and prevent contamination during production.

Raw materials are pumped into a mixing head following the proportions specified in Table 1. The temperature of the polyol and TDI are controlled with heat exchangers, ensuring optimal processing conditions for the reaction. The dosing system used for injection operates at high pressures, with TDI being injected at pressures of up to 100 bar. The foam mixture is poured onto a pour plate, which disperses it onto a moving conveyor. The conveyor is equipped with a controlled speed and is lined with kraft paper sheets on the top, bottom, and sides to shape and contain the foam during expansion. The reaction mixture attains a creamy consistency, and in approximately 2 min, the foam expands. The foam is conveyed for 6 to 15 min following expansion to allow for curing. In a tunnel, the foam is shaped into large blocks with approximate dimensions of 1.2 m in height and 2 m in width, which are cut into lengths of 60 m. Following expansion and shaping, the foam blocks are transferred to a storage area to complete curing under controlled conditions, allowing residual heat dissipation generated by the exothermic PU formation reactions. During curing, the blocks must dissipate heat until thermal stabilization is achieved, reaching room temperature. Due to the high temperatures generated during the curing of the blocks, it is essential to maintain an effective and operational detection and extinguishing system. This system includes various equipment and infrastructure, such as the emergency generator and fire pumps, which are considered within the scope of this study. Once fully cured, the foam is ready for its intended application. Also considered in this study are the solvent, acetone, and mesamoll, which are auxiliary materials used for cleaning tools employed in the process.

The production of PUF is based on the fundamental chemistry of PU formation, where the isocyanate group exhibits high reactivity towards hydroxyl (OH)-containing compounds. This includes functional groups present in water, alcohols, carboxylic acids, amines, ureas, and urethanes. Consequently, numerous concurrent reactions can occur. The preparation of PU typically involves the reaction of isocyanates with polyols, as depicted in Figure 1, to form urethane bonds. Also, another main reaction occurs between the isocyanate group and water, yielding amines and CO₂, the latter crucial for the cellular structure of the PUF [9]. The water used in the mixing and reaction stage undergoes treatment via reverse osmosis to ensure purity and suitability for use as a blowing agent in the reaction with isocyanate.

2.3. Inventory Analysis

The inventory data used to produce the flexible PUF is summarized in

Table 3. Inventory data to produce 1 kg of flexible PUF.

Parameter	Unit	Flexible PUF
		Conventional	Bio-Based
Inputs
Raw materials
Polyol mix	g	645	584
584Bio-polyol	g	-	103
TDI	g	321	271
Water	g	260	21.9
Catalysts	g	1.84	2.09
Silicones	g	6.44	7.54
Pigments	g	0.20	10.3
Ancillary materials
Kraft paper	g	9.27	9.43
Lubricating oils	g	0.0006	0.0007
Mesamoll	g	0.0038	0.0039
Solvent	g	0.034	0.035
Acetone	g	0.023	0.023
Nitrogen	g	4.50	4.60
IBC	g	0.705	0.722
Metallic drums	g	0.237	0.243
Wooden pallets	g	0.784	0.803
Energy
Electricity	kWh	0.0244	0.0250
Diesel	g	0.049	0.049
Natural gas	MJ	0.0159	0.0159
Outputs
Flexible PUF	g	1000	1000
Air emissions
CH4	mg	0.000735	0.000735
CO2	mg	1.090	1.090
CO	mg	0.712	0.712
NOx	mg	1.790	1.790
NMVOC	mg	0.0366	0.0366
SOx	mg	0.0231	0.0231
PM10	mg	0.0122	0.0122
PM2.5	mg	0.0122	0.0122
N2O	mg	0.00671	0.00671
NH3	mg	0.000392	0.000392
Wastes
Aqueous suspension	g	0.221	0.226
Amines	g	0.136	0.140
Pigment	g	0.0237	0.0243
Waste oil	g	0.0006	0.0007
Polyol mix	g	1.51	1.55
Kraft paper	g	9.27	9.49
IBC	g	0.705	0.722
Metallic drums	g	0.237	0.243
Wooden pallets	g	0.784	0.803

. Apart from the air emissions, the data considered are average primary data obtained from an industrial plant in Portugal. This data was complemented with information from the Ecoinvent database version 3.9.1 [23], particularly to produce the raw and ancillary materials required to produce the flexible PUF. In turn, data on the production of polyether polyol, polypropylene glycol and amine, and the HDPE tank were sourced from Industry Data 2.0 database, Carbon Minds cm.chemicals database [24] and BUWAL250 database, respectively.

Regarding air emissions, both from natural gas burning, used for producing hot water in the boiler for the raw material heat exchangers, and from diesel-burning in the emergency generator and fire pumps, including methane (CH₄), nitrogen oxides (NO_x), carbon monoxide (CO), non-methane volatile organic compounds (NMVOC), sulphur oxides (SO_x), particulate matter (PM₁₀ and PM_2.5), dinitrogen oxide (N₂O), and ammonia (NH₃), were calculated based on emission factors from the joint European Monitoring and Evaluation Programme (EMEP)/European Environment Agency (EEA) [25], and data on CO₂ were calculated based on the emission factors from the Portuguese Environment Agency [26].

Data on electricity generation in the Portuguese grid were sourced from the Ecoinvent database. Waste is assumed to be incinerated, disposed of in non-hazardous landfills, recycled, or reconditioned for industrial packaging. Data on waste treatment was taken from the Ecoinvent database. In this case, the cut-off procedure was adopted, meaning no environmental burdens from the recycling or reconditioned processes were allocated to the systems analysed.

Table 4. Transport profile for 1 kg of flexible PUF.

	Distance (km)	Transport Mode
Raw and ancillary materials
Polyol mix	324	Lorry, Euro 5 (16–32 t)
	9691	Sea, container ship
Bio-polyol	45	Lorry, Euro 5 (16–32 t)
	4844	Sea, container ship
TDI	230	Lorry, Euro 5 (16–32 t)
	3849	Sea, container ship
Catalysts	2564	Lorry, Euro 5 (16–32 t)
	10,673	Sea, container ship
Silicones	4222	Lorry, Euro 5 (16–32 t)
Pigments	90	Lorry, Euro 5 (16–32 t)
	6930	Sea, container ship
Kraft paper	210	Lorry, Euro 5 (16–32 t)
Waste
Aqueous suspension	288	Lorry, Euro 5 (16–32 t)
Amines	306
Pigment	328
Waste oil	60
Polyol mix	275
Kraft paper	17
IBC	55
Metallic drums	55
Wooden pallets	3

presents the type of transport used for different raw and ancillary materials and waste, and the estimated average distances travelled.

2.4. Impact Assessment

The impact assessment was performed based on the ReCiPe 2016 midpoint method [27] for the following impact categories: global warming (GW), fossil resource scarcity (FRS), mineral resource scarcity (MRS), terrestrial acidification (TA), freshwater eutrophication (FE), marine eutrophication (ME), and ozone formation related to human health (OF-HH). The impact categories were selected based on multiple criteria, including the relevance to the evaluated systems and the robustness level of their characterization models. Additionally, the selection process ensured the inclusion of impacts associated with resource depletion and air and water pollution. As explained by Huijbregts et al. [27], these characterization models quantify how specific emissions contribute to each impact category, enabling their aggregation into comparable environmental indicators.

2.5. Sensitivity Analysis

To ensure that the properties of flexible PUF outlined in Table 2 were similar to those of the reference PUF, the incorporation of bio-polyol used was only 10%. The decision to use this percentage was based on the team’s experience and numerous reports in the literature, which consider 20% as the maximum percentage of replacement of crude-derived polyols by bio-polyols. For that reason, a sensitivity analysis was conducted to assess the potential impact of this bio-polyol maximization on the LCA results of flexible PUF formulations. However, this analysis should be regarded as an approximation, and the results should be interpreted cautiously. This is because, at the industrial scale, additional modifications to the bio-based formulations may be necessary, as certain properties, e.g., density and IFD, could be affected depending on the source of bio-polyol.

3. Results and Discussion

3.1. General Environmental Impacts

Table 5. Total impacts for 1 kg of flexible PUF.

Impact Category	Unit	Flexible PUF
		Conventional	Bio-Based
GW	kg CO2 eq	3.87 × 100	3.65 × 100
FRS	kg oil eq	1.75 × 100	1.60 × 100
MRS	kg Cu eq	6.85 × 10−3	6.44 × 10−3
TA	kg SO2 eq	1.15 × 10−2	1.07 × 10−2
FE	kg P eq	6.45 × 10−4	7.13 × 10−3
ME	kg N eq	1.26 × 10−4	1.56 × 10−4
OF-HH	kg NOx eq	6.86 × 10−3	6.75 × 10−3

shows the total impacts of producing 1 kg of flexible PUF using fossil-based and bio-based formulations (Table 1). The conventional flexible PUF presents the highest impacts in GW, FRS, MRS, and TA, except for FE and ME. This shows the potential of bio-based PUF flexible PUF to reduce climate-related impacts compared to its fossil-based counterpart.

Reducing fossil resource dependency directly lowers GW impacts of bio-based flexible PUF, supporting the EU’s carbon-neutral goals under the European Green Deal. While specific greenhouse gas (GHG) reduction values for materials like flexible PUF are not defined, transitioning to bio-based alternatives can contribute to the broader industrial emissions reductions targeted for 2050 [28].

Notably, the OF-HH impacts of bio-based flexible PUF are very close to those of conventional PUF. For FE, the impacts of bio-based flexible PUF are by far the highest, compared to the FE impacts of conventional flexible PUF, i.e., 91% higher. For ME, the impacts of bio-based flexible PUF are also higher than those of the conventional flexible PUF but at a lower magnitude (19% higher). The highest FE for bio-based flexible PUF resulted mainly from using phosphorus-based fertilisers during soybean cultivation. This soybean is the bio-polyol feedstock, as mentioned in Section 2.1.

As regards the ME, the bio-based formulation exhibits a 23% contribution associated with the bio-polyol. This is primarily linked to nitrate emissions to water associated with vegetable oil refining and soybean production. These trade-offs illustrate that bio-based materials mitigate GW but raise concerns about eutrophication impacts.

3.2. Hotspots Analysis

Figure 4 shows the contribution of each raw and ancillary material, transport, electricity, and direct emissions to the production of flexible PUF. Depending on the impact category under study, the polyol mix, bio-polyol, and TDI are the main hotspots.

Regarding the conventional-based flexible PUF, the polyol mix and TDI are the main contributors, accounting for 50% and 45% of the total GW impact, respectively. For FRS, the polyol mix is also the hotspot, with a 61% contribution, followed by TDI at 36%. Despite the partial substitution of fossil-based polyols by bio-polyols, in the case of the bio-based flexible PUF, the polyol and TDI remain the main contributors for these impact categories: the polyol mix and TDI account for 47% and 40% of the total GW impact, respectively, whilst for FRS impact, 60% is associated with the polyol, followed by TDI at 33%.

These results reflect the dependence of flexible PUF on fossil resources, particularly in the synthesis of polyols and isocyanates derived from petroleum-based intermediates. The environmental impacts associated with the polyol mix are primarily attributed to producing propylene oxide, a key precursor during the polyether polyol synthesis [29]. The TDI impacts are predominantly driven by the production of its key intermediates, particularly 2,4-dinitrotoluene and phosgene, which rely on nitric acid and toluene as primary raw materials. In the case of TDI, GW impact mainly results from N₂O and CO₂ emissions generated during the production of nitric acid and toluene. Furthermore, the FRS impacts are primarily attributed to crude oil and natural gas extraction, which serve as essential energy sources in the upstream production processes of toluene, and ammonia, the latter being a precursor for nitric acid.

In conventional and bio-based flexible PUF scenarios, the MRS and TA impact categories are primarily driven by TDI and polyol mix production. For conventional flexible PUF, TDI accounts for approximately 60% of the total MRS and TA impacts, while the polyol mix contributes 36% and 29% to MRS and TA, respectively. A comparable pattern is observed in the bio-based PUF scenario. TDI is responsible for 54% of the MRS impact and 55% of the TA impact, followed by the polyol mix with contributions of 38% to MRS and 27% to TA. These results highlight the substantial mineral resource requirements associated with the production of flexible PUF, particularly during the intermediate stages of TDI production. This is due to the extraction of mineral resources such as sulphur, used in sulfuric acid production for the synthesis of 2,4-dinitrotoluene, and the raw materials required for polyol production, including the extraction of potash salt and limestone used in the propylene oxide process [29]. These findings apply to both conventional and bio-based flexible PUF, underscoring the resource-intensive nature of their upstream production processes.

In the conventional flexible PUF formulation, eutrophication impacts are predominantly driven by TDI, which accounts for 88% of the total impacts in the FE. This is mainly due to emissions associated with producing key TDI intermediates, particularly phosgene and 2,4-dinitrotoluene, and electricity consumption. As regards the ME, the polyol mix is the leading contributor with 37%, followed by TDI at 35% and catalysts at 23%. The high ME impact of catalysts is primarily attributed to ammonium ion emissions released into water systems during their production.

In the bio-based flexible PUF scenario, eutrophication impact profiles shift significantly. Bio-based polyol is responsible for over 92% of the total impact of FE. This is mainly driven by phosphorus and chemical oxygen demand emissions in the wastewater generated during vegetable oil refining. In the ME category, the bio-based polyol contributes to 23% of total impacts, mainly due to nitrate emissions linked to wastewater from vegetable oil refining and the agricultural phase of soybean cultivation.

In the bio-based flexible PUF, a notable shift in ME impact is also observed, with the bio-based polyol emerging as a new dominant contributor. Nevertheless, catalysts maintain a substantial contribution of 22% to ME, like the conventional fossil-based formulation, indicating that while the primary hotspot changes, eutrophication impacts from catalyst production remain relevant across both systems.

The polyol mix primarily influences the OF-HH impact category and TDI, with contributions of 45% and 37%, respectively, in the conventional PUF, and slightly reduced contributions of 39% and 32% in the bio-based PUF. These impacts are attributed mainly to the release of ozone precursors, particularly NOx, during upstream production processes such as the synthesis of propylene oxide in polyether polyol production and the upstream production of toluene and nitric acid used in the manufacture of 2,4-dinitrotoluene, as well as the synthesis of phosgene in the TDI production chain. Transport contributions remain below 10% across all impact categories for conventional and bio-based PUF, except for the OF-HH category. Transport accounts for 13% and 15% of the total impact in the conventional and bio-based scenarios, respectively. This contribution is mainly attributed to NOx emissions from the combustion of heavy fuel oil during the maritime transport of polyether polyols (Table 4), accentuating the negative side effects of logistical operations in the PUF supply chain.

3.3. Sensitivity Analysis

Figure 5 presents the sensitivity analysis results, illustrating the environmental impacts of increased bio-polyol content in the formulation from 10% to 20% by weight relative to conventional polyol (100% fossil-polyol mix). When comparing the two bio-based flexible PUF scenarios (one with 10% and the other with 20% bio-polyol), the latter shows a reduction in the GW, FRS, and MRS impacts by 3.0%, 8.0%, and 3.2%, respectively. Conversely, the FE and ME impacts increase significantly by 48% and 16%, respectively. These increases are primarily driven by emissions associated with the intensification of soybean cultivation, including nutrient runoff (phosphorus and nitrogen compounds) and leaching processes linked to fertilizer use and land management practices.

The variation in the OF-HH and TA impacts is minimal (<1%), indicating negligible sensitivity to the increase in bio-polyol content. These findings are consistent with those observed in the base case of the bio-based flexible PUF, reinforcing that increasing the proportion of bio-polyol generally contributes to impact reductions in climate- and resource-related categories (GW, FRS, MRS), but leads to environmental trade-offs in water-related eutrophication impacts (FE and ME).

Despite the doubling of bio-based polyol content, which led to a proportional reduction in the polyether polyol mix, the relatively modest decrease in GW impact (3%) can be attributed to elevated GHG emissions associated with soybean cultivation, namely diesel combustion in agricultural operations. Moreover, the increase in soybean production and its intensive input requirements, including fertilizers, also contribute to the elevated MRS and TA impacts.

Interestingly, while diesel use in soybean cultivation still plays a role in the FRS category, the overall FRS impact decreases by 8.0%. This suggests that the reduction in fossil-based polyol consumption, previously identified as a major hotspot for fossil resource use, outweighs the increased diesel demand in the agricultural stage. Therefore, the transition towards a higher bio-polyol content offers potential benefits for fossil resource depletion. However, these are partially offset by the environmental burdens inherent to bio-based feedstock production. Therefore, future research and sustainability strategies should consider the potential benefits of more environmentally responsible agricultural practices in bio-polyol feedstock production. While not assessed in this study, improved cultivation methods (potentially involving optimized fertilizer application or integrated soil fertility management) could further enhance the environmental performance of bio-based polyols and increase their potential as sustainable alternatives in PUF production.

3.4. Comparison with Other Studies

It is not possible to directly compare the total impacts obtained in this study with those reported in previous LCA studies on flexible PUF because different impact assessment methods were applied, and the system boundary may include distinct inputs and outputs.

Nevertheless, Europur [10] reported results for GW using characterisation factors from the Intergovernmental Panel on Climate Change reports. Therefore, a comparison is only feasible for this impact category. Even though the EPD of Flexible PUF released by Europur [10] only considers the use of conventional polyols, it is important to compare it with the results obtained in our study regarding the conventional flexible PUF to validate the methodology followed. This EPD identifies long-chain polyol precursors and TDI as the main contributors to environmental impacts in flexible PUF production, collectively accounting for over 85% of the total impact across environmental categories. Specifically, the GW is 3.44 kg CO₂-eq per kg of TDI fossil-based flexible PUF with a density ranging from 18 to 25 kg/m³, and 3.45 kg CO₂-eq per kg of TDI fossil-based flexible PUF with a density ranging from 35 to 40 kg/m³. Our result (3.87 kg CO₂-eq/kg conventional PUF under study) aligns closely with Europur’s EPD, highlighting raw materials, especially polyols and isocyanates, as the main contributors to environmental impacts in flexible PUF production. Furthermore, it also demonstrates the relevance of using real industrial data besides validating the methodology followed in our present study. To facilitate comparison,

Table 6. Comparison of GW values for flexible PUF.

Source	System Description	Density (kg/m3)	Impact Assessment Method	GW (kg CO2-eq/kg Foam)	Hotspots
Europur [10]	Conventional flexible PUF (fossil polyol)	18–25	EF v3.1	3.44	Polyol, TDI
Europur [10]	Conventional flexible PUF (fossil polyol)	35–40	EF v3.1	3.45	Polyol, TDI
Present study	Conventional flexible PUF (fossil polyol, industrial data)	22–26	ReCiPe 2016 midpoint	3.87	Polyol, TDI
Present study	Bio-based flexible PUF (partial substitution with bio-polyol, industrial data)	29–32	ReCiPe 2016 midpoint	3.65	Bio-polyol, polyol, TDI

summarizes the GW results reported by Europur [10] and those obtained in this study.

4. Conclusions

This study provides a cradle-to-gate LCA comparing fossil-based and bio-based flexible PUF formulations at the industrial scale. The incorporation of bio-polyols derived from vegetable oils led to environmental improvements in GW (6%), FRS (9%), and MRS (6%), supporting the transition toward lower dependence on petroleum-based feedstocks. However, increases in FE (91%) and ME (19%) were observed, mainly due to agricultural inputs, including fertilizer use and nutrient runoff associated with vegetable oil production.

The main hotspots identified were the polyol mix, TDI, and bio-polyol production (only for bio-based flexible PUF), as precursor synthesis and agricultural stages largely determine the environmental profile of flexible PUF. Sensitivity analysis indicated that increasing the bio-polyol content to 20% provided limited additional reductions in climate change impacts while intensifying eutrophication effects, underscoring the need for a balanced integration of bio-based materials.

Integrating industrial-scale data in this study strengthens the robustness and representativeness of the results, ensuring that the environmental indicators reflect real manufacturing conditions and provide relevant insights for sustainable decision-making within the flexible PUF industry.

The transition to bio-based and alternative PUF formulations remains essential for improving the sustainability of this sector, but continued efforts are necessary to enhance agricultural practices, optimize bio-polyol synthesis, and reduce emissions along the value chain. These combined measures will certainly minimize environmental trade-offs and accelerate the shift from fossil-based to renewable feedstocks.