Househeating Pulse
EU Heat-Pump Market Intelligence

Brand signal · 6 min read · Published 2026-05-31

Belgium 2026: which heat-pump brands still lead on R32, not R290

Belgium stands out in 2026 because its heat-pump market has not flipped to R290 as quickly as much of Europe. The article should quantify the brand mix and show which makers are keeping R32 alive.

gray and brown building under blue sky
Photo by Marius Badstuber on Unsplash

Belgium’s 2026 heat-pump market in numbers

Belgium is still an outlier in 2026 because R32 accounts for 13,935 EPREL-listed heat-pump models versus just 537 for R290, a gap of 13,398 models and 21.97 percentage points across the indexed market snapshot used here (market_index_snapshot).

That is the central signal from the current EPREL-based dataset: the catalog is still overwhelmingly HFC-led, not propane-led. On the same snapshot date, the sample covers 60,989 models from 777 manufacturers, with an overall average SCOP of 4.55 (market_index_snapshot). If you want the broader baseline, Househeating Pulse’s market index snapshot and manufacturer directory show how unusual that refrigerant balance now looks against the wider market.

The refrigerants actually present in the declared listings are also highly concentrated. R32 alone appears on 13,935 listings, while R410A appears on 1,896, R410 on 10, R410a on 49, and R290 on 537; the rest are tiny residual codes such as R134A at 2 and a string of one-off declarations including R23, R420A, R423A, R421A, R411A, R35, R422A, R413A, R425A, R419A, R33, R417A, R332 and R41OA, each with 1 listing (refrigerant_universe). EPREL code quality plainly still includes variant spellings and legacy entries, which is worth checking in the refrigerants reference before drawing technology conclusions from raw labels alone.

One limitation matters up front: the corpus supplied here does not include a Belgium-only EPREL country cut with manufacturer-by-refrigerant breakdown. So the article can quantify the refrigerant mix and leading brands in the supplied 2026 snapshot, but it cannot prove a Belgium-only brand table beyond that corpus.

How far Belgium is from Europe’s R290 shift

The surprise is not that R290 exists in the dataset; it is how small it still is. In the current snapshot, R32 represents 22.85% of all listed models, while R290 represents 0.88% (computed from 13,935 and 537 out of 60,989) (market_index_snapshot). That leaves R32 ahead of R290 by 21.97 percentage points (market_index_snapshot).

By count, the market remains even more lopsided than the percentages suggest: there are roughly 26 R32 listings for every one R290 listing in the sample (market_index_snapshot). That is the opposite of the “all-propane now” narrative often implied in market commentary.

The same snapshot reports a natural refrigerant share of 3.27% overall (market_index_snapshot). Because the declared refrigerant table is still dominated by R32 and older HFC codes, Belgium’s lag looks less like a narrow technology exception and more like a market still carrying a very large installed product pipeline around A2L HFCs. For context on where national markets sit relative to one another, the 32-country comparison dashboard and the country index are the relevant reference points.

The user asked for an EU benchmark for R32 and R290 and Belgium’s distance from that benchmark. The corpus does not provide an EU-wide refrigerant-share snapshot separate from this market index, nor a Belgium-only split against a defined EU aggregate. So that part cannot be answered precisely from the supplied data.

Which brands still keep R32 alive

The leading manufacturers in the supplied snapshot are concentrated at the top. Daikin Europe N.V. leads with 14,668 models and 24.05% share, followed by Mitsubishi Electric Europe B.V. with 5,575 and 9.14%, Johnson Controls Hitachi with 5,207 and 8.54%, Bosch Thermotechnik GmbH with 3,602 and 5.91%, and Ariston SpA with 2,618 and 4.29% (brand_share). Those five brands alone account for 51.93% of all listed models in the dataset (brand_share).

That concentration matters because any delayed refrigerant transition will be driven disproportionately by the portfolio choices of a handful of suppliers, not by the long tail. Readers can inspect those portfolios directly via Daikin Europe N.V., Mitsubishi Electric Europe B.V., Bosch Thermotechnik GmbH, and Ariston SpA, or browse the full heat-pump catalog.

But there is a hard data gap here. The corpus does not contain Belgium-market manufacturer refrigerant splits, so it cannot answer which specific brands in Belgium have an R32 share, R290 share, or absolute number of R32 models in Belgium. It also cannot identify which Belgian-market brands meet the threshold of at least 50% of portfolio on R32. That requires a brand-by-country-by-refrigerant table that is not present in the supplied JSON.

The manufacturers most exposed to the R32 lag

What the corpus can show is which manufacturers matter most in absolute catalog weight. Beyond the top five, the next tier is Atlantic at 1,516 models and 2.49% share, Vaillant at 1,195 and 1.96%, BDR Thermea at 925 and 1.52%, General HVAC Solutions Euro at 921 and 1.51%, and Panasonic Marketing Europe at 894 and 1.47% (brand_share).

Add the top ten together and they account for 60.88% of all models in the snapshot (brand_share). In other words, the remaining 767 manufacturers share just 39.12% of listings (market_index_snapshot; brand_share). That is a highly concentrated structure by model count, and it strengthens the case that Belgium’s slower shift is at least partly a portfolio-composition story.

However, the user asked which brands account for the largest absolute number of R32 models. That cannot be answered from this corpus because the necessary brand-level refrigerant counts are unavailable; the supplied brand_detail probes for Daikin, Mitsubishi Electric and Vaillant all failed (brand_detail). The same missing detail prevents a ranked list of the most R32-exposed manufacturers.

What the refrigerant mix says about technology choice and efficiency

The efficiency side is clearer than the refrigerant side. The overall snapshot’s average SCOP is 4.55 (market_index_snapshot). Several large brands sit above that level: Bosch at 4.69, Ariston at 4.66, Gree at 4.65, Ferroli at 4.64, Riello at 4.60, and Gorenje at 4.67 (brand_share). Others are below it, including Daikin at 4.44, Atlantic at 4.38, BDR Thermea at 4.37, Panasonic at 4.30, Hitachi at 4.18, and Toshiba Carrier at 3.93 (brand_share).

So the lag in the refrigerant transition cannot be reduced to a simple “higher efficiency explains R32” story from the data provided. Among the largest manufacturers, efficiency performance is mixed, not uniformly better for the brands likely to be carrying legacy HFC-heavy portfolios. If anything, the range of 3.93 to 4.69 average SCOP across the top 15 shows that refrigerant choice and seasonal efficiency are related only loosely in the current listing base (brand_share).

That also means buyers should not treat refrigerant code as a proxy for best performance. The more useful workflow is to cross-check R32 models in the catalog, R290 models in the catalog, and the top SCOP leaderboard or air-to-water SCOP leaderboard.

Why Belgium remains an outlier in the European transition

The strongest explanation available from this corpus is structural, not technological. The market snapshot is dominated by a few very large brands, with the top five controlling 51.93% of listings and the top ten controlling 60.88% (brand_share). At the same time, the refrigerant base remains dominated by 13,935 R32 listings, while R290 is still only 537 listings and natural refrigerants overall just 3.27% of the sample (market_index_snapshot).

That combination points to a slow transition path whenever major manufacturers continue to carry large R32 portfolios through EPREL. It does not prove that Belgium alone has the same exact split as the aggregate market snapshot; again, the corpus lacks a Belgium-only refrigerant-and-brand cut. But it does support the article’s core signal: the remaining R32 footprint is not being preserved by niche suppliers. It is being preserved by market concentration around very large manufacturers whose listed portfolios still dominate the sample (brand_share; market_index_snapshot).

For readers tracking the broader policy and compliance angle, the refrigerant table also matters because R32 is listed with a 2027-01-01 phase-out date in the reference table, while R290 is a natural refrigerant with GWP 0 (refrigerant_universe). That raises the stakes for any market still leaning heavily on R32 in 2026.

Sources

  • market_index_snapshot — Househeating Pulse · Market Index v1, computed from EPREL Public API. Snapshot: 2026-05-31.
  • refrigerant_universe — IPCC AR6 GWP table; EU Reg. 2024/573 phase-out schedule; EPREL declared codes. Snapshot: 2026-05-31.
  • brand_share — EPREL Public API · brand-share aggregation. Snapshot: 2026-05-31.
  • brand_detail — (probe failed — data unavailable). Snapshot: .

Continue reading