Mattresses and furniture are a huge potential source of PU for recycling, but they are complex products made using many materials. As we move into 2022, extended producer responsibility and changing rules are focusing minds in the EU industry
Extended producer responsibility makes manufacturers responsible for the waste created by their products at the end of their useful life. It is proving to be a powerful motivator of innovation within the EU. It is now common to see the formation of consortia of companies and organisations to determine the best way to deal with the tsunami of end-of-life mattresses that the EU generates every year.
Describing it as a tsunami is not hyperbole – the volume is staggering. Marcel Moeller, global marketing & sustainability director at Dow Polyurethanes, explained the situation to participants at EuroPUR’s conference in November last year. ‘Within the EU, 30m mattresses on average are discarded every single year,’ he said. ‘Most of these go into landfill or are incinerated. If they were piled up, they would be 680 times the height of Everest.’
Scrap it
Bert Haltermann, corporate manager for sustainable innovation at Recticel, told the meeting the latest EuroPUR figures for end-of-life mattresses indicate that the EU generates about 200kT/year of polyurethane which is available for recycling. Currently about half ends up in landfill, with 45% incinerated with energy recovery. Just 5% is mechanically recovered.
But that is only volume from end-of-life mattresses. The volume of scrap resulting from the manufacture of mattresses and upholstered furniture is larger, and the scrap from that furniture once it is discarded even greater. ‘The bed, furniture and mattress makers generate about 310kT/year trim foam, all of which is mechanically recycled,’ Halterman said. ‘But the really big volumes are in end-of-life furniture upholstery. EU processors generate about 660kT/year of this, and about 50% is incinerated and 50%, landfilled. In contrast, the automotive sector produces about 130kT/year of PU foam scrap, which is mechanically recycled, 50% incinerated and the rest landfilled.’
If all the figures are totted up, it amounts to a pool of 1.3mT/year of waste PU foam in the EU for processing. That’s a huge volumetric problem. ‘I estimate that if all the mattress foam were landfilled, then Europe would need twice the volume of the pyramid of Cheops in Egypt,’ Halterman said. This equates to about 2.6m m3 being buried in the EU each year.
Dow’s Moeller explained that attacks on several fronts will be required if the problem is to be solved. While it is important to look at the technology, an ecosystem of companies and organisations to support the circular economy needs to be in place. But compared to linear value chains, circular chains are complex.
‘We need to start creating those circular loops,’ he said. ‘We need to step up to that challenge collectively. Today, pressure is building on PU because it is perceived as being non-recyclable. We are all in this together – it is an industry problem.’
Yet it is not a problem solely for the foam makers, and some big guns in the downstream furniture manufacturing sector are taking it very seriously. For example, in December 2019 IKEA said its goal was to become climate positive and circular by 2030. ‘[We want to reduce] greenhouse gases emissions by more than the total IKEA value chain emits, while growing our business,’ the company said. ‘We will not rely on carbon offsetting, and we will work within our value chain.’
Moeller described the EU’s waste recycling targets as very ambitious. ‘Different countries are at different stages of progress,’ he said. ‘France, for example, is among the front runners.’
The EU’s Green Deal foresees that by 2050 the EU will be ‘a fair and prosperous society, with modern, resource efficient and competitive economy where there are no net emissions of greenhouse gases in 2050 and economic growth is decoupled from resource use’.
But this target for 2050 is not an isolated aspiration: there are also ambitious interim targets. By 2030, the EU wants to cut greenhouse gases by 50% of the 1990 level, and by the middle of 2022, it may extend the emissions trading system into a number of additional sectors, and price carbon effectively throughout the economy.
High profile
Recycling large, easily identifiable products such as mattresses will help to meet those targets, and the stakes are high. The PU industry is actively working to meet these standards, as several speakers at the EuroPUR meeting explained.
For example, Haltermann said that Recticel is working with the French recycling coordination organisation EcoMobilier and automotive seating products company Tesca on the ValPUMat project. This programme is designed to extract value from end-of-life mattresses by using mechanical recycling to develop foam-based products, cars and acoustic building insulation. The project started in 2018, and will last for three years. During that time, it is expected to generate 10 new applications for foam from end-of-life mattresses.
One of the project’s successes has been to make acoustic foam for buildings, generators and compressors, as well as the passenger compartment of cars. Conventional techniques shred the foam, turning it into a fleece bonded by isocyanate and steam. ValPUMat has developed a new material: a fleece bonded by fibres in a continuous process. Recticel is running a commercial plant to make it in France, and a pilot plant in Belgium.
A second development from the project is SilentWall. This is an 80kg/m3 foam bonded to plasterboard, designed to help insulate rooms from noise.
Recticel is also part of the Puresmart consortium, funded by the EU’s Horizon 2020 programme. This was set up in early 2018 with the goal of developing a complete circular life cycle for polyurethane, turning it into a ustainable material. As well as reducing the amount of waste generated, the Puresmart programme should also help to generate 20 full time jobs for every 6kT/year of PU recycled.
Its core goals are to divert up to 220 kT/year PU waste from landfill, and reduce the carbon intensity of PU foam by at least 30%. It fits with the EU’s goal that member states should reuse and recycle 55% of municipal waste, including mattresses and furniture, by 2025 and the amount of municipal waste going to landfill to 10% by 2030. Once again, large, bulky objects like mattresses are good targets.
The Puresmart programme has three pillars: smart design, smart sorting and smart chemolysis. The design pillar is based on developing new polyurethanes with a greater thermoplastic character that could be easier to recycle than conventional fully thermoset materials. By the end of 2020, Halterman said, a number of candidates based on triazolinedione were being studied. These can be used as co-monomers with polyols in formulations to create covalent adaptable polyurethane (CAPU).
The manufacture of these raw materials will be scaled up by Weylchem, and then used by Recticel in its range of products. This foam should retain the mechanical properties of conventional polyurethane foam while being much easier to recycle because of their thermoplastic nature, according to project information.
As we reported in our June/July 2020 issue, the consortium is developing a smart sorting process with Austrian company Redwave. The aim is to fractionate post-consumer PU waste from bedding and furniture, and this will be fed into chemical recycling processes developed by groups at the KU Leuven and the University of Castile La Mancha. The separation technique uses machine learning algorithms. Halterman described it as a breakthrough for PU foam recycling.
The consortium has three patent applications for processes in its smart chemolysis pillar. The programme members have built on a lab method chemolysis that gives high purity polyols at a high yield after screening a number of agents and separation methods. The group is investigating the most effective way to scale this up to a semi-industrial process that will generate polyols and amines for conversion into diisocyanates.
Sort that waste
Dow is part of another consortium that has made high-profile advances recently, winning an award from Chemical Week magazine on the way. Moeller explained the company’s plans to sell Renuva brand polyols made from post-consumer mattress waste collected under France’s Eco-Mobilier scheme. They will be generated at Orrion Chemicals’ Orgafoam site in France, using technology developed by German company H&S.
Turning the value chain from the straight line of produce–sell–scrap into a circle where the scrap is used in the raw materials for the production phase will have an immediate impact on the waste problem, Moeller claimed. ‘Only one industrial scale plant will recycle around 200,000 mattresses/year,’ he said. ‘The Renuva system is scalable and decentralised, with huge potential,’ he said. ‘It is a demonstration that PU can be recycled and that recovered polyols have value.’
The chemical recycling process converts waste PU into polyols with up to 50% recycled content. ‘This polyol will be used as a high-value constituent in new flexible foam for mattresses, and it could be used in rigid foams for thermal insulation,’ he said. ‘It is important here to understand that the value stays very high so the properties remain in place.’
There is no need to segregate PU foam types, he said, and any kind of PU foam can be recycled, as long as it is clean and dry. The first plant is under construction, and should soon be operational.
Applications are already in the pipeline, with Vita announcing in October 2020 that it plans to use polyols from the project in its foams. At the time, Vita’s CEO Ian Robb said: ‘We are committed to being pioneers in the development of a circular economy within our industry, where post-consumer flexible PU foams will be recycled into new raw materials for use in our manufacturing process to produce new, quality foams.’
Vita expected to start using polyols at its ICOA plant in Crancey, France in early 2021. It also intends to work with mattress brands in France and across Europe to promote the use of foam made with recycled polyols in the new mattress market. It also plans to look for other applications for the material.
In-process scrap
Polish flexible foamer Dendro’s recycling activities first came to prominence in 2013, when it announced it was working with H&S to recycle in-process scrap at its facility in Rogozno, about 350km west of Warsaw. Dendro now trades as Ikano Industry, and is still a key supplier to Ikea, being owned by the Kamprad family which controls the Swedish furniture giant. It supplies about 2.5m mattresses to Ikea, according to Michal Soltsinski, Ikano’s head of PU foam production.
The company started looking at post-industrial waste in 2015. ‘We invested,’ Soltsinski said. ‘It took three years from the feasibility study while we searched for the best available technique with a focus on chemical recycling.’
The company now has a 7m³ reactor that makes polyol from cut-offs and waste foam from industrial production. These include both HR and conventional foam, he said, and the waste is free from fillers, melamine and other flame retardants, colorants and pigments. As the demand for repolyol is growing, the company is now investing in a 15m³ reactor.
The only criteria for foam to be recycled are that it must be both clean and dry. ‘It needs good preparation before we use it,’ he said. ‘Otherwise, if something sticks to the foam it will be found in the reactor. We use a range of different polyols in different formulations. These include standard, visco, repolyol, natural oil and CO₂ based polyols. hey can all be recycled. The manufacturing cost of the repolyol is about 50% of the cost of standard polyol, based on a market price of €1.5/kg.’
Soltsinski estimates that since 2015 when the project started, Ikano has generated 12.5kT of repolyol, equivalent to the capacity of 500 road tankers, and consumed 5.5kT of trim foam. That represents, he said, about 250 truckloads.
Ikano is making repolyol with a specification very similar to standard polyol in terms of hydroxide number and water content. ‘Its viscosity is higher, and we worried about that at the start, but it works well in production,’ he said. Repolyol can substitute for between 20% and 40% of virgin polyol. Ikano has also tried it at 100% and made a 50m long block. That foam had lower physical properties, but he believes it will still have applications.
Initially, there were concerns that some of the polyol would be reused many, many times, and that this could have an effect on finished properties.
However, Soltsinski explained, at an addition level of 20-40%, this does not seem to be a problem.
Ikano decided on that upper addition level of 40% because the foam had to meet Ikea’s demanding compression set specifications. ‘Ikea sells many of its mattresses pressed and rolled with the volume reduced, so the rolled height of the foam is 10% of the original height,’ he said. ‘For low density foam of about 25kg/m³, the maximum would be about 25%.’
After 2017, the company began to look at the possibility of using post-consumer waste, too. ‘Post-consumer waste has some biological content, and this could be hard to sell into the flexible foam market where there can be a lot of near personal contact with flexible foam products,’ Soltsinski said. ‘To get around this, we looked at glycolysis as a method of producing polyols for rigid foams.’ The company has already successfully used the polyols to make rigid sandwich panels.
‘In parallel, we tried acidolysis,’ he added. ‘We did not think that this would be as promising, but today it is the main process we use with material from the post-consumer waste stream.’
Of course, biological content is not the only factor that makes post-consumer waste much more demanding to reprocess than post-industrial waste. The composition is far more unpredictable, perhaps because the foam has been combustion modified in some way, he said. As well as flame retardants and other additives, it can incorporate materials such as glue, textiles and latex. ‘This is very different to post-industrial waste, where we know exactly what is in our foam,’ Soltsinski said.
Ikano has already scaled this work up from the lab into a 700kg reactor, which is sufficiently large to make quantities for testing on processing machines. ‘The foam parameters are OK,’ he said.
‘There is a drop in resilience and hardness, but tensile strength and extension at break increase.’
As well as physical properties, Ikano tested the foam from a consumer safety perspective at a number of external institutes. The foams were also tested against Ikea IOS-MAT 0010,0054, which covers the migration of flame retardants, lead, organotin compounds, phthalates, TDA, MDA and other products. Foams were also tested for microbial growth on the on the foam to ensure they would be safe for sale to consumers.
Lesson learnt
The need for a continuous source of post-consumer waste led to the team making other discoveries, Soltinski said. They were given some mattresses that had been returned to Ikea as faulty products,. This gave them the opportunity to look at the dismantling process, which proved challenging.
‘We realised that this needed a lot of knowledge,’ he said. ‘It also made us realise that there must be some kind of future design for recycling. It should be much easier than it is now, and we have started looking at that with Ikea.’
Their work on dismantling highlighted several factors that would make the process simpler. ‘You should not mix PU with latex, for example,’ he said. ‘There should be 100% polyurethane foam or 100% latex to help with the separation.’ Foam should not be attached to the cover, either, as this is difficult to separate effectively.
Armed with that knowledge, Ikano is now cooperating with RetourMatras, a Netherlands-based mattress recycling company. It has the collection, dismantling and logistical expertise that will be required for Ikano to make repolyol from post-consumer waste into an important input stream of material for the manufacture of future consumer products.
