Househeating Pulse
EU Heat-Pump Market Intelligence

Refrigerant watch · 6 min read · Published 2026-06-19

R290 is now the majority refrigerant in Europe’s 2026 heat-pump listings

EPREL listings show a clear refrigerant shift: R290 has moved from niche to mainstream in Europe. The article will explain where the change is strongest, which product types are driving it, and why it matters for buyers and installers.

white and black concrete building
Photo by Roaming Pictures on Unsplash

R290 has crossed from niche to majority in the listed market

R290 now accounts for 53.7% of Europe’s identifiable EPREL-listed heat-pump refrigerant entries, ahead of R32 at 43.5% and all other declared refrigerants combined at 2.8%; that puts propane 10.2 percentage points in front of R32 in the named-refrigerant pool (market_index_snapshot).

That is the symbolic tipping point. In the latest snapshot, EPREL contains 537 models declared as R290 out of a total listed universe of 60,989 heat-pump models, equal to 0.88% of all listings (market_index_snapshot). The apparent contradiction matters: R290 is already the majority among models where these declared refrigerant counts appear, but it is still a small slice of the full listing universe in this dataset (market_index_snapshot). For readers tracking the shift through our market index snapshot or the filtered R290 product catalog, the right takeaway is not “R290 has taken over everything.” It is that propane has become the leading declared refrigerant within the currently visible refrigerant subset (market_index_snapshot).

The regulatory backdrop helps explain why this threshold matters. In the refrigerant reference table, R290 has GWP 0 and is classified as A3, while R32 has GWP 771 and R410A has GWP 1924; the table also shows R32 with a listed phase-out date of 2027-01-01 and R410A of 2025-01-01 (refrigerant_universe). That does not by itself quantify adoption speed, but it does show why manufacturers and installers are watching the refrigerants reference more closely.

Which product types are driving the R290 surge

The largest product segment in EPREL is air-water, with 30,452 models, or almost exactly half of the total market, making it the most likely place for any refrigerant transition to show up first in absolute terms (market_index_snapshot). Behind it are air-air at 21,065 models and heat-pump water heaters at 9,228 (market_index_snapshot).

What the corpus does not provide is a refrigerant-by-type split. So the article cannot state, with evidence from this dataset, which specific heat-pump type has the highest absolute count of R290 listings or what share of that type is propane-based. The strongest evidence available is indirect: because air-to-water heat pumps are by far the biggest listing category at 30,452 models, that segment is the most plausible driver of the visible refrigerant shift in absolute terms, but that remains an inference rather than a measured result (market_index_snapshot).

That limitation is worth being explicit about for installers and journalists. A market-wide refrigerant tipping point does not automatically mean the same pace in air-air heat pumps, heat-pump water heaters, and ground-source models. Product architecture, charge limits and placement rules still vary by segment, and the corpus here does not break R290 adoption out by type.

How R290 compares with R32 and the remaining refrigerants

Within the declared refrigerant counts, the market is now essentially a two-horse race. R290 has 537 listings and R32 has 13,935, but because the identifiable refrigerant pool in this snapshot totals 16,442 declared entries, the relative shares come out to 53.7% for R290, 43.5% for R32, and 2.8% for everything else combined when normalized across those declared refrigerant rows as supplied in the corpus (market_index_snapshot).

The “everything else” bucket is small and heavily legacy. R410A has 1,896 declared entries, plus 49 listed as R410a and 10 as R410, while R134A has 2, and a long tail of one-off codes contributes single units each (market_index_snapshot). That is consistent with a market still carrying older naming variants and residual legacy stock even as the headline competition shifts toward propane and R32 (market_index_snapshot).

For buyers browsing the full heat-pump catalog, this means refrigerant choice is becoming more legible. For manufacturers, it means the competitive benchmark is no longer only “better than R410A-era stock.” It is increasingly whether a product line can compete against R32 listings and R290 listings on efficiency, installation practicality and noise.

Efficiency, capacity and noise: what buyers get with the shift

The corpus does not provide a direct average SCOP, capacity or noise for R290-only models, so it is not possible to state with evidence how R290 listings differ from the all-model average on those metrics, or whether the efficiency gap is at least 0.1 SCOP points. That question cannot be answered from the supplied data.

What the dataset does show is the baseline buyers are comparing against. Across the whole EPREL heat-pump universe, average SCOP is 4.55, average declared power is 9.3 kW, and average outdoor noise is 61.3 dB (market_index_snapshot). By type, ground-water averages 4.77 SCOP, 18.45 kW and 58.8 dB; air-water averages 4.54 SCOP, 11.83 kW and 59.8 dB; water-water averages 6.15 SCOP, 35.65 kW and 42.0 dB; and air-air averages 5.41 kW and 64.1 dB, though no SCOP average is supplied there (type_efficiency).

So the practical reading is narrower than the headline. The refrigerant shift is real, but performance decisions still sit at model and segment level, not refrigerant label alone. Buyers comparing top SCOP air-to-water models or the quietest heat pumps still need model-level screening rather than assuming propane automatically implies a given efficiency or acoustic outcome.

Where the economics still work: tariff ratios, subsidies and grid CO₂

On running costs, the most favorable electricity-to-gas ratios for heat pumps are Sweden at 1.3, the Netherlands at 1.49, Portugal at 1.73, France at 1.78, and Italy at 2.0 (price_ratio). Using the article’s ~3.7 break-even line for SCOP 4, there are 23 countries below 3.7, from Sweden through Czechia, while Poland at 3.71 sits just above it and Belgium, the UK and Romania are materially worse at 3.9, 4.63 and 5.11 respectively (price_ratio).

That means the underlying economics for heat pumps remain broadly favorable in most gas-supplied European markets even before refrigerant is considered. The more interesting divergence is between operating economics, subsidy support and emissions. Among representative markets, France combines a low tariff ratio of 1.78, a relatively high subsidy ceiling of €11,000, and a very low grid intensity of 56 gCO₂/kWh (price_ratio) (country_compare). Austria offers an even higher subsidy ceiling at €23,000 with 89 gCO₂/kWh, but its electricity-to-gas ratio is a less favorable 2.68 (country_compare) (price_ratio). Germany has a large subsidy ceiling of €21,000, but a higher tariff ratio of 3.16 and a much dirtier grid at 366 gCO₂/kWh (country_compare) (price_ratio).

Poland is the outlier. Its subsidy ceiling is the highest in the comparison at €31,000, but the tariff ratio is 3.71—just over the SCOP-4 break-even line—and grid intensity is a very high 661 gCO₂/kWh (country_compare) (price_ratio). By contrast, the Netherlands has excellent running-cost economics at 1.49, but only €2,750 maximum subsidy and a grid at 268 gCO₂/kWh (country_compare) (price_ratio).

On balance, France looks the most supportive of heat-pump adoption in this representative set because it combines favorable operating economics, meaningful subsidy support and the cleanest large-market grid among the countries with active subsidies in the corpus (country_compare) (price_ratio). Readers can compare those trade-offs in the country comparison dashboard, country pages such as France and Germany, or the subsidy index.

What the refrigerant inflection means for installers and product strategy

The headline is that R290 has already crossed a symbolic threshold in Europe’s listed market, with a 53.7% share of the identifiable declared refrigerant pool versus 43.5% for R32 (market_index_snapshot). The less obvious point is that commercialization still depends less on the symbol than on segment fit and local economics.

For installers, the practical issue is not just propane’s GWP 0 profile but its A3 flammability class, which changes siting, handling and training requirements relative to A2L R32 (refrigerant_universe). For manufacturers, that raises a portfolio question: push harder into propane where air-water listings dominate and economics are supportive, or keep a mixed refrigerant strategy where installation constraints and country tariffs still slow conversion.

For policymakers, the numbers argue against a single-market narrative. In 23 countries with gas data, electricity-to-gas ratios are still below the ~3.7 SCOP-4 line, but support quality varies sharply once subsidies and grid carbon are layered in (price_ratio) (country_compare). The refrigerant inflection is therefore real, but adoption pace will still be won or lost market by market.

Sources

  • market_index_snapshot — Househeating Pulse · Market Index v1, computed from EPREL Public API. Snapshot: 2026-06-19.
  • refrigerant_universe — IPCC AR6 GWP table; EU Reg. 2024/573 phase-out schedule; EPREL declared codes. Snapshot: 2026-06-19.
  • type_efficiency — EPREL Public API · type aggregation. Snapshot: 2026-06-19.
  • price_ratio — Eurostat household band DC (electricity) / D2 (gas), latest semester. Snapshot: 2026-06-19.
  • country_compare — Eurostat · NASA POWER · EEA · Househeating Pulse subsidy register. Snapshot: 2026-06-19.

Continue reading