Insulation is one of the most contested product categories in construction-EPD work. Five very different chemistries — mineral wool, glass wool, EPS, XPS, polyurethane, and a growing share of bio-based materials — compete in the same buying decisions. Each has its own PCR, its own data quirks, and its own typical verifier pool. This article is a quick orientation for a manufacturer starting their first EPD process.
The chemistries and their PCR landscape
Mineral wool (stone wool, rock wool)
The most heavily declared insulation type by EPD volume. PCRs are mature, the production data is fairly consistent across manufacturers, and the verifier pool is deep. The dominant impact driver is the energy used in melting the basalt/slag feedstock — typically 60–75% of the cradle-to-gate GWP. Production-data discipline around furnace energy is the single most important contributor to a defensible EPD.
Glass wool
Similar to mineral wool in PCR coverage and verifier availability. The recycled-cullet content is the headline lever — manufacturers using a high share of recycled glass generally land with better GWP numbers. Verifiers will scrutinise the allocation method for the cullet (closed-loop vs open-loop, system expansion vs cut-off) — pick the method with your consultant before the LCA model is built.
Expanded polystyrene (EPS)
The bead-expansion and steam-moulding steps are the main energy hotspots, alongside upstream styrene production. EPS EPDs are well-developed in Europe (via IBU, Environdec) and in North America (via UL). Watch the biogenic carbon treatment — EPS uses pentane as the blowing agent, but the underlying styrene supply chain has its own GHG profile that verifiers will probe.
Extruded polystyrene (XPS)
Historically the most carbon-intensive of the mainstream insulations because of the HFC blowing-agent legacy, though the modern HFO/CO₂-based formulations have closed the gap considerably. EPDs for XPS need to explicitly declare the blowing agent and its global-warming potential; if you've recently switched formulations, the EPD has to reflect the current product, not the historical one.
Polyurethane (PUR / PIR)
Complex upstream supply chain — isocyanates, polyols, and flame retardants each have their own LCA tail. The verifier work is heavier here than for the silicate-based insulations. Most PUR/PIR EPDs come from larger manufacturers because the data-collection lift is non-trivial for an SME.
Bio-based (wood fibre, hemp, cellulose, cork)
The fastest-growing segment by EPD volume in 2025–2026. The key methodological question is the treatment of biogenic carbon — most modern PCRs require biogenic-carbon accounting (separate from fossil GWP), and the declared number can swing dramatically depending on the method. For wood fibre and hemp, the upstream agricultural/forestry data is often the biggest unknown; for cellulose, the recycled-content allocation is the headline question.
The data you'll need to collect
Across all insulation chemistries, the LCA model needs:
- Production volume by product, by plant, for the reference year
- Energy consumption (electricity, thermal — split by fuel)
- Raw-material inputs with supplier names and transport modes
- Process-waste streams and their disposal/recovery routes
- Product properties: declared thickness, density, R-value (insulation EPDs typically declare per unit area at a defined R-value, not per unit mass)
- End-of-life scenario assumptions
The end-of-life scenario is more sensitive in insulation than in most product categories — recyclability assumptions can move the cradle-to-grave GWP by 10–20%. Verifiers will challenge optimistic assumptions, so document the basis carefully.
The program operators you'll encounter
For European-market insulation, IBU dominates the German market, Environdec is widely used internationally, and MRPI is the de facto Dutch route. France's INIES via FDES is mandatory for RE2020-regulated buildings. For North American markets, UL and NSF are the main routes; the EPA Label Program will increasingly drive product-specific declarations as the GWP thresholds tighten through 2027.
The competitive context
Insulation is one of the few sectors where buyers actively compare EPDs across competing products at the procurement stage — architects often want to see two or three options side by side. This makes EPD quality particularly important: a sloppy declaration with wide uncertainty ranges, missing modules (B and C), or unfavourable PCR choices reads badly when placed next to a well-prepared competitor's EPD. The cost of a "good" EPD in insulation is largely a cost of marketability.
Two specific recommendations for insulation manufacturers in 2026: (1) Include modules B and C in your declaration even where the PCR only requires A1–A3 — buyers are increasingly looking at full life-cycle. (2) Be ready for renewals — your first EPD will likely need updating sooner than 5 years if you switch a feedstock or a blowing agent, and several major buyers now ask for EPDs less than 3 years old.