Europe’s Chemical Risk Reset Lands on the Construction Supply Chain

Europe’s Chemical Risk Reset Lands on the Construction Supply Chain

Empa’s contribution to the European Partnership for the Assessment of Risks from Chemicals reached its halfway point this summer with a working prototype of something the construction supply chain has never previously had access to: a structured method for pricing the safety of a substance before anyone has specified it.

The Swiss Federal Laboratories for Materials Science and Technology co-lead the Safe and Sustainable by Design sub-task within PARC, a seven-year, roughly €400 million partnership running to 2029 that connects more than 200 institutions across 29 countries, with the European Chemicals Agency, the European Food Safety Authority and the European Environment Agency all inside the governance structure.

The stated ambition is modest in phrasing and considerable in consequence. As Empa researcher Bernd Nowack of the Technology and Society department puts it, “We want to develop tools that can be used to identify the potential dangers of chemicals to humans and the environment right at the start of the development process.”

For a sector that specifies epoxy coatings on potable water mains, fluoropolymer membranes on roofs, admixtures in structural concrete and flame retardants in cladding, the timing is more commercially loaded than the research framing suggests. Around 350,000 chemicals are manufactured and used worldwide, with new entrants arriving continuously, and construction consumes a disproportionate share of them in long-life assets that cannot easily be recalled.

The industry has already paid twice for substances that passed the technical specification and failed the toxicological one decades later, and it is about to be asked, through the recast Construction Products Regulation, to disclose in machine-readable form exactly what is in every product it places on the European market. A toolbox that can flag a problem at formulation stage is not an academic curiosity in that context; it is the difference between a supplier holding a defensible position in 2035 and holding a remediation provision.

Briefing

• Empa co-leads the Safe and Sustainable by Design sub-task of PARC, the EU’s €400 million chemical risk assessment partnership, which runs to 2029 and involves more than 200 institutions across 29 countries.
• Initial versions of the SSbD toolbox have been validated against bisphenol A, a substance with direct construction relevance through epoxy resin coatings, pipe liners and tank linings.
• The European Commission confirmed in April 2026 that it will not proceed with a full revision of REACH, pushing regulatory pressure into sector-specific instruments and design-stage assessment instead.
• ECHA’s Risk Assessment Committee adopted its final opinion on the universal PFAS restriction in March 2026, with the Socio-Economic Analysis Committee due to conclude by the end of the year.
• Empa has extended the framework beyond chemicals to materials, completing a full SSbD assessment of graphene-based materials now entering commercial concrete and asphalt formulations.

The Regulatory Centre of Gravity Has Shifted Away From REACH

The most significant market development sitting behind this research is one the Commission announced rather than proposed. On 27 April 2026, Environment Commissioner Jessica Roswall told the European Parliament’s ENVI Committee that the long-delayed overhaul of REACH would not go ahead, concluding that reopening the regulation was not appropriate at a moment when industry needed legal certainty and predictability.

The revision had been repeatedly postponed since 2022 and had received a negative opinion from the Regulatory Scrutiny Board in September 2025. In its place the Commission will pursue targeted simplification through comitology, amending REACH annexes via secondary legislation without full parliamentary involvement, alongside a renewed focus on enforcement.

Suppliers reading that as a reprieve are reading it wrongly. The pressure has not been removed, it has been redistributed into instruments that are already law and already biting. Regulation (EU) 2024/3110, the recast Construction Products Regulation, entered into force in January 2025 and became generally applicable on 8 January 2026, embedding a construction-specific Digital Product Passport into CE marking and the EU Construction Products Database.

Its data fields extend well beyond load-bearing performance to recycled content, disassembly potential, expected service life and the presence of substances of very high concern. Obligations crystallise product family by product family through delegated acts, which means most high-volume categories such as concrete, steel and insulation will see requirements land in the second half of this decade rather than immediately. The practical effect is that manufacturers now have a defined window to get their bill of materials, supplier declarations and test certificates into a queryable structure before disclosure becomes mandatory rather than optional.

Alongside that sits the universal PFAS restriction, submitted to ECHA in January 2023 by authorities in Denmark, Germany, the Netherlands, Norway and Sweden, and covering a class estimated at more than 10,000 substances. ECHA’s Risk Assessment Committee adopted its final opinion on 2 March 2026, concluding that existing measures are insufficient and EU-wide action is needed.

The Socio-Economic Analysis Committee agreed its draft opinion in March, ran a final consultation to 25 May 2026 covering construction products explicitly among the sectors under review, and is expected to adopt its final position by the end of the year. Both committees have recommended removing or significantly reducing many of the derogations in the original proposal. Legal analysis suggests the Commission is unlikely to adopt any restriction before the third quarter of 2027, with market effects landing from 2029. Sealants, membranes, coatings, fire-fighting installations and machinery components all sit inside that perimeter.

Why Bisphenol A Was the Right Test Case for Infrastructure

The choice of bisphenol A as the first substance run through the toolbox looks like a laboratory convenience and functions as something closer to a stress test of the construction supply chain. Nowack is direct about the reasoning: “In collaboration with other institutions, we applied the tools to bisphenol A (BPA).” He adds that “A great deal is already known about the risks and effects of this chemical. This allowed us to test how well our models work.”

BPA is a foundation monomer for epoxy resins, and epoxy is the dominant resin system in cured-in-place pipe lining, the trenchless technique used to rehabilitate water mains from 100mm to 2,400mm diameter without excavation. It also appears in tank linings, industrial flooring, protective coatings on steel and concrete, and adhesives.

That matters because the compliance clock has already started. The revised Drinking Water Directive (EU) 2020/2184 introduced BPA as a parameter with a value of 2.5 micrograms per litre, and Member States were required to have measures in place to meet the new parametric values by 12 January 2026. Separately, Commission Regulation (EU) 2024/3190 bans BPA in food contact materials, with narrow exemptions including liquid epoxy resins used as heavy-duty varnishes and coatings, and transitional provisions running to July 2026 and January 2028 depending on product type.

Research published in Environmental Science & Technology in January 2025 by a Purdue University team examining new epoxy drinking water pipe liners found that laboratory results scaled to pipe diameters between 100mm and 900mm indicated that the smallest liners would push BPA concentrations above EU and WHO drinking water limits, and recommended volatile and semi-volatile organic compound testing before liners are returned to service. European water associations have gone further, arguing that BPA emissions from the polymer, concrete and steel industries where epoxy coating is common should be consolidated and addressed, and that BPA should be removed from pipes, valves and fittings in hot water systems where release is materially higher than at ambient temperature.

None of this makes trenchless rehabilitation the wrong answer for Europe’s ageing water networks, and the sector has demonstrable alternatives in BPA non-intent formulations, polyurethane systems and inner film barriers. What it does establish is that a utility awarding a framework for pipe rehabilitation in 2026 is making a chemistry decision as much as a civil engineering one, and that the toolbox Empa is helping to build is aimed squarely at the moment when that decision is still cheap to change.

The Legacy Bill That Sets the Investment Case

The commercial argument for design-stage assessment rests on what the alternative has cost. Asbestos remains the reference case, and it is still generating expenditure two decades after the EU ban. Directive (EU) 2023/2668 lowered the occupational exposure limit tenfold, from 0.1 to 0.01 fibres per cubic centimetre as an eight-hour time-weighted average, with Member States required to transpose the bulk of the provisions by 21 December 2025 and remaining technical measurement requirements by December 2029.

The Commission published implementation guidelines in December 2025 aimed specifically at construction, renovation and maintenance. Because the Renovation Wave will disturb asbestos already in situ across millions of pre-1992 buildings, the practical burden falls on contractors delivering exactly the retrofit programmes Europe needs for its energy performance targets.

The PFAS figures are larger and more recent. Peer-reviewed work published in Environmental Science: Processes and Impacts this year, developed through the cross-border Forever Lobbying Project, estimated legacy remediation costs across the EU27 at approximately €37 billion over 20 years, or €1.8 billion annually, for removing previously emitted long-chain PFAS from hotspot sites and drinking water.

The emerging scenario, which targets shorter and ultra-short chain compounds including trifluoroacetic acid across wastewater effluent and biosolids, was estimated at roughly 20 times that figure, at around €100 billion per year. A separate European Commission report in 2026 projected soil remediation and drinking water treatment costs of up to €80.2 billion per year and health costs of up to €39.5 billion per year across the period to 2050, depending on how much preventative action is taken. The researchers’ own conclusion, that reducing use at source is more efficient than treating contamination afterwards, is the entire economic thesis behind SSbD compressed into a sentence.

For infrastructure owners and their insurers, those numbers reframe substance selection as a balance sheet exposure with a multi-decade tail. For European materials manufacturers, they also describe an opportunity. A supplier that can demonstrate, with documented modelling rather than assertion, that its formulation carries no comparable liability profile holds a commercial advantage in tender evaluation that did not exist five years ago and will be increasingly hard to replicate.

Graphene, Admixtures and the Limits of Chemical-Era Tools

Empa’s most instructive extension of the toolbox moved from chemicals to materials, applying the full SSbD framework to graphene-based materials in work published in Environment International by Fiorella Pitaro, Stefan Seeger and Nowack. The exercise was designed to test the framework rather than to deliver a verdict on graphene, and it exposed precisely the gaps that matter to construction.

Assessment models built for discrete chemicals produced significant variation when applied to a material family, and in some cases proved inapplicable. Under certain modelled production scenarios, worker exposure exceeded the derived no-effect level, which is a finding about occupational controls at manufacturing scale rather than about finished products in service. The team also had to treat graphene as a family rather than a substance, spanning pristine single-layer graphene, graphene oxide and reduced graphene oxide, each with distinct hazard profiles despite performing similar industrial functions.

That distinction has immediate commercial weight because graphene admixtures are no longer speculative in the built environment. Concretene, developed with the University of Manchester’s Graphene Engineering Innovation Centre, reports a 25 to 30 per cent increase in compressive strength in laboratory testing and an average of 17 per cent in field trials, translating into a potential 10 to 15 per cent reduction in concrete-related carbon emissions depending on application.

Arup took an equity stake in the venture, Black Swan Graphene supplies material under a partnership involving an equity swap, and a £3 million financing round co-led by Molten Ventures with LocalGlobe closed in October 2024 with product certification explicitly named as a use of funds. UK government support has backed precast work with piling manufacturer Roger Bullivant and sleeper development with the Global Centre for Rail Excellence, alongside progress in paints, polymers and asphalt.

Certification is the operative word. A product entering structural concrete and railway sleepers needs a defensible position on human and environmental exposure across manufacture, placement, service life and eventual demolition or recycling. Empa’s finding that the existing modelling toolkit struggles with material families is therefore not a discouraging result for the graphene supply chain. It identifies the specific methodological work that has to be completed for advanced materials to clear procurement and insurance thresholds at scale, and it does so while the industry is still small enough to absorb the answer.

Building the Practitioner Base That Industry Will Have to Hire

The harder constraint Nowack identifies is interpretive rather than computational. “We can assess and quantify the risks,” he explains, with the difficulty lying in what those figures mean when almost no empirical data exists for a genuinely new substance. His group works with estimates, models and probabilities, and he is convinced that a distinct professional category will be required to operate the framework credibly, describing a future need for specially trained SSbD practitioners in both industry and the public sector. The scale of the undertaking is not trivial even within a partnership of PARC’s size, and he notes that “Our part of PARC is as big as an entire “standard” EU research project.”

The template being examined comes from sectors that already work this way. “SSbD – at least the ‘safe by design’ part – is already standard practice in the pharmaceutical and agricultural industries,” Nowack explains, adding that “If a drug is not safe for humans, it is not developed any further.”

The researchers are now studying how pharmaceutical and agrochemical development processes can be transferred into other industrial sectors, and are pushing for closer collaboration with manufacturers on the reasoning that regulatory bodies are not where the relevant decisions are made. As Nowack puts it, “SSbD must be taken into account as early as possible in the development process of a chemical – and these development processes mainly take place in industry.”

For construction chemical suppliers, admixture producers and coating formulators, that signals a skills requirement arriving alongside the digital product passport obligations, and firms that build the capability early will find themselves shaping the assessment methods their competitors later have to comply with.

Positioning Before the Next Substance Cycle Begins

The pattern worth acting on is the one that connects the Commission’s decision to leave REACH alone with the arrival of the construction Digital Product Passport. Horizontal reform has stalled, so scrutiny is arriving instead through product-level disclosure, drinking water parameters, occupational exposure limits and class-based restrictions, each with its own timetable and each requiring the same underlying evidence about what a product contains and what it releases.

Manufacturers who treat the DPP as a data architecture problem rather than a compliance form will find that the same substance-level records satisfy tender questionnaires, EPD verification under EN 15804, insurer due diligence and eventual REACH annex amendments without duplicated effort.

Asset owners have a parallel opportunity in how they write specifications and frameworks. Requiring documented substance disclosure and design-stage safety assessment as an evaluation criterion, rather than a warranty, moves the risk to the party best placed to manage it and rewards the suppliers investing in exactly the capability PARC is trying to standardise.

The European materials sector has a genuine competitive position to defend here, because the assessment infrastructure is being built inside European research institutions with European regulators in the governance structure, and firms engaging with it now will be operating the tools rather than reacting to them. That is a considerably better place to be than the one the industry occupied when the last generation of problem substances came due.

Key Industry Questions

1. What does Safe and Sustainable by Design actually require a manufacturer to do? The SSbD framework, developed by the European Commission’s Joint Research Centre and published as a Commission Recommendation in 2022, sets out a staged assessment covering hazard, human health and environmental risk in production and use, environmental sustainability across the life cycle, and social and economic dimensions. It is currently voluntary rather than a legal obligation. In practice it asks a formulator to run safety and sustainability screening in parallel with technical development rather than after it, and to document the comparison between candidate formulations. PARC’s toolbox exists to make that practical by assembling validated tools for each step, so that companies are not required to build assessment methodology from scratch.
2. Is SSbD compliance mandatory for construction products sold in the EU? Not directly. SSbD remains a voluntary framework, but the evidence it generates increasingly overlaps with obligations that are mandatory. The recast Construction Products Regulation requires environmental and substance information in the Declaration of Performance and, through delegated acts, in the Digital Product Passport. The Drinking Water Directive sets enforceable parametric values for substances including bisphenol A. REACH restrictions under Annex XVII are binding once adopted. A manufacturer that has run an SSbD assessment will already hold most of the underlying data those instruments demand, which is why the framework is best understood as compliance infrastructure rather than an optional exercise.
3. How exposed is the water sector to bisphenol A in epoxy pipe liners? Exposure varies with liner diameter, resin formulation, curing method and water temperature. The parametric value of 2.5 micrograms per litre under Directive (EU) 2020/2184 has applied since January 2026, giving utilities a defined compliance threshold. Published research indicates that smaller-diameter cured-in-place liners present the greatest risk of exceedance, and that curing conditions and hardener ratios materially affect residual monomer levels. Practical mitigation is well established: specifying bisphenol A non-intent formulations, requiring inner barrier films, mandating post-installation flushing protocols and commissioning organic compound testing before return to service. The technique itself remains a sound answer for network rehabilitation when the chemistry is specified deliberately.
4. When will the PFAS restriction actually affect construction products? ECHA’s Risk Assessment Committee adopted its final opinion in March 2026 and the Socio-Economic Analysis Committee is expected to finalise its position by the end of 2026, after which the dossier passes to the European Commission. Legal commentators expect a Commission proposal to amend REACH Annex XVII no earlier than the third quarter of 2027, with restrictions on some categories taking effect from 2029. Construction products were specifically covered in the final consultation. Market effects are already visible ahead of any legal deadline, with manufacturers reporting supply insecurity for PFAS-containing components and some suppliers withdrawing products voluntarily rather than carrying reformulation risk.
5. What should manufacturers be doing about the Digital Product Passport now? The immediate work is connecting systems rather than generating new information. Batch records, test certificates, bills of materials and supplier declarations usually already exist inside the organisation but sit in formats that were never designed to be queried externally. Because product-family obligations arrive progressively through delegated acts, most manufacturers are not currently non-compliant, which makes this the cheapest period in which to restructure that data. Firms holding verified EN 15804 environmental product declarations already have a substantial part of the environmental dataset. The gap is usually substance-level traceability through the supply chain, which requires supplier engagement and takes longer than internal system work.
6. Does design-stage assessment slow down materials innovation? The pharmaceutical and agrochemical precedent suggests the opposite once the process is established. Those sectors screen for safety at the earliest development stages and abandon candidates that fail, which concentrates investment on formulations capable of reaching market. Construction chemicals have historically deferred that screening, which produces late-stage failures and, in the worst cases, post-market withdrawal after commercial deployment. The transitional cost is real, principally in skills and modelling capability, but the effect on capital efficiency is favourable. Reformulating a coating at laboratory scale is considerably cheaper than reformulating it after it has been specified across a national asset portfolio.
7. Why is graphene assessment relevant if the material is barely deployed? Graphene-enhanced admixtures are moving through certification into structural concrete, precast piling and railway sleepers, with Arup, Black Swan Graphene and UK government programmes among the backers. Assessment work completed while deployment volumes remain modest is far more useful than assessment attempted afterwards. The Empa study also served a broader purpose by testing whether tools built for single chemicals work for material families, and the answer was that significant adaptation is required. That finding applies equally to nanocellulose, engineered fibres, novel binders and other advanced construction materials now approaching commercialisation.
8. What does the abandoned REACH revision mean for long-term planning? It removes a source of uncertainty about the shape of the horizontal framework while leaving the direction of travel unchanged. The Commission will amend REACH annexes through comitology and has signalled stronger emphasis on enforcement, particularly around registration dossier quality and imported goods. Planning assumptions should therefore centre on sector-specific instruments already in force, the construction Digital Product Passport, drinking water parameters, occupational exposure limits and class-based restrictions under Annex XVII, rather than on a future consolidated reform. For most manufacturers this means the compliance workload arrives incrementally and predictably rather than in a single legislative event.

Strategic Takeaways

1. Substance-level data architecture is now a competitive asset rather than a compliance overhead, because the same records satisfy Digital Product Passport obligations, EPD verification, tender questionnaires and insurer due diligence simultaneously.
2. The decision not to reopen REACH does not reduce scrutiny of construction materials; it distributes that scrutiny across product-level disclosure, drinking water parameters and class-based restrictions with staggered and independent timetables.
3. Remediation economics have decisively overtaken prevention economics, with European PFAS clean-up estimates running from €1.8 billion annually for legacy contamination to roughly €100 billion annually under broader scenarios, which makes design-stage screening the rational commercial default.
4. Advanced materials entering structural applications, graphene admixtures foremost among them, will need assessment methods that existing chemical-era models cannot yet deliver, and the suppliers funding that methodological work will shape the certification standards their competitors must later meet.
5. A shortage of trained SSbD practitioners is emerging as the binding constraint on adoption, creating a recruitment and training requirement for construction chemical suppliers, admixture producers and coating formulators well before regulation compels it.