Quick answer. China bans HCFC-141b in rigid foam from January 1, 2026 (premixed systems) and July 1, 2026 (all remaining sectors, including spray foam), under MEE Announcement No. 28 (2025). The three replacement routes are cyclopentane (cheapest agent, needs ATEX-rated equipment), HFO-1233zd(E)/1336mzz(Z) (near drop-in, non-flammable, roughly 6-10x cyclopentane's cost), and water/CO2 blowing (minimal retrofit, higher density and lambda penalty). Most panel and appliance lines are converting to cyclopentane; spray crews and capex-constrained plants are leaning HFO.
For buyers sourcing rigid PU or PIR systems out of China, this isn't a distant policy footnote — it's a supplier qualification question due within the next few quarters.
Past the cutoff, any TDS or COA still naming HCFC-141b as the blowing agent is a red flag, not a rounding error. This guide compares the three conversion routes, the formulation parameters each one moves, and what a plant retrofit actually requires.
What China's 2026 HCFC-141b Ban Actually Requires
Under Announcement No. 28 (2025), production of premixed polyether and polyurethane systems using HCFC-141b as the blowing agent is prohibited from January 1, 2026, covering everything except on-site spray-applied foam.
From July 1, 2026, the ban extends to spray-applied polyurethane systems and closes out every remaining rigid foam sector — construction spray, appliance insulation panels, cold-chain storage panels, and industrial pipe-in-pipe foaming.
The underlying reason hasn't changed since the original Montreal Protocol listing. HCFC-141b carries an ozone depletion potential (ODP) of 0.11 and a 100-year global warming potential in the 700-800 range.
HFO alternatives run near-zero ODP and single-digit or sub-1 GWP. The US EPA's SNAP program reached the same conclusion years earlier, listing HCFC-141b as unacceptable for rigid PU/PIR boardstock once lower-ODP substitutes became commercially available.
Practically, this means any supplier contract or TDS still referencing HCFC-141b needs a hard revision date attached to it, not a promise that a reformulation is underway.
Three Conversion Routes at a Glance
Every plant converting off HCFC-141b is choosing between three physical or chemical blowing mechanisms. Each one moves a different set of formulation and capital variables.
The right choice depends on your line type, target lambda, and how much capex is available before the deadline.
| Property | Cyclopentane (n-/iso-/cyclo blends) | HFO-1233zd(E) / 1336mzz(Z) | Water / CO2 (all-water) |
|---|---|---|---|
| ODP | 0 | ~0 (effectively zero) | 0 |
| 100-yr GWP | <20 | <1 | 1 (CO2 is the GWP reference gas) |
| Flammability | Flammable; flash point commonly cited between -50°C and -6°C depending on isomer blend | Non-flammable (ASHRAE 34 Class 1) | Non-flammable agent; reaction exotherm still needs managing |
| Typical initial λ | ~21-23 mW/m·K | ~19-21 mW/m·K | ~24-27 mW/m·K |
| Typical 10-yr aged λ | ~24-26 mW/m·K | ~21-23 mW/m·K | ~28-31 mW/m·K |
| Core density for equivalent R-value | Baseline, sometimes +2-4 kg/m³ vs HCFC-141b | Equal or slightly lower than HCFC-141b baseline | +15-25% density typically needed |
| Drop-in to existing HCFC-141b dosing equipment | No — requires ATEX-rated retrofit | Largely yes — same low-pressure dosing hardware | Yes, but mixhead residence time and exotherm control need retuning |
| Indicative agent cost | Lowest — commodity hydrocarbon pricing | Roughly 6-10x cyclopentane on a $/kg basis | Lowest reagent cost, offset by higher polyol/isocyanate use per m³ |
Formulation Changes Each Route Forces on Your System
Cyclopentane systems
Cyclopentane's gas-phase thermal conductivity runs higher than HCFC-141b's. Matching the same aged lambda usually means a finer, more uniform cell structure.
That takes tighter catalyst balance (gel vs. blow) and a silicone surfactant re-selected for cyclopentane's lower polyol solubility — it phase-separates faster than HCFC-141b in the resin blend.
Flame retardant loading typically goes up too. HCFC-141b's chlorine content contributed passive flame suppression that hydrocarbons don't.
Formulators commonly raise TCPP or a comparable halogenated phosphate ester by several pphp to hold the same B2 (DIN 4102) or GB 8624 class, then retest rather than assume the old rating carries over.
Dosage runs roughly 8-14 pphp depending on target free-rise density. Premix polyol shelf life also needs re-verification, since cyclopentane's higher vapor pressure builds drum pressure faster than HCFC-141b did.
HFO systems
HFO-1233zd(E) and 1336mzz(Z) are close to drop-in on catalyst package and surfactant — both are closer to HCFC-141b in polarity and solubility than cyclopentane is.
Most reformulations land within 1-3 pphp of the original HCFC-141b dosage, needing only minor catalyst rebalancing to match the same cream/gel/tack-free profile.
The bigger risk is isocyanate index drift. HFOs have a different vapor pressure curve, which shifts foam rise timing on fast-index spray or panel lines.
That timing needs re-mapping on the actual production mixhead, not just the lab bench.
Water/CO2 systems
Raising water content from a trace co-blow level (0.5-2 pphp) to an all-water system (typically 3-5 pphp) generates more urea hard-segment content.
That raises Tg and can make the foam more brittle at low temperature — a real concern for cold-chain panels cycling repeatedly to -25°C.
The added exotherm from water/isocyanate chemistry also runs hotter than physical blowing, raising scorch and core discoloration risk on thicker pours.
Catalyst timing and mold or line-speed parameters need re-tuning, not just the water dosage.
Plant and Equipment Retrofit Reality Check
This is where cyclopentane and HFO diverge hardest. Cyclopentane is flammable — LEL around 1.1% v/v, UEL around 8.7% v/v for typical blends.
That flammability pulls in ATEX/Ex-proof requirements that HCFC-141b lines never needed. HFO systems are non-flammable and mostly reuse existing low-pressure dosing hardware.
| Requirement | Cyclopentane line | HFO line |
|---|---|---|
| Ex-proof (ATEX Zone 1/2) motors, pumps, controls | Required | Not required |
| Nitrogen inerting/purge on storage & day tanks | Required | Not required |
| LEL gas detection network | Required, typically alarming at 20% LEL | Not required |
| Blowing agent storage | Outdoor bunded tank farm, intrinsically safe instrumentation | Standard indoor drum/IBC, same footprint as HCFC-141b |
| Grounding/static bonding on transfer lines | Required | Standard practice sufficient |
| Reported incremental capex | Tens to several hundred thousand USD, scaling with line throughput | Minimal — existing HCFC-141b dosing equipment is usually reusable |
Occupational exposure is the other gap most teams underestimate. OSHA's Table Z-1 permissible exposure limits carry no enforceable PEL specific to cyclopentane.
The 1989 rulemaking that proposed a 600 ppm TWA was later vacated, so plants generally apply the ACGIH/NIOSH-recommended level as internal practice rather than a regulatory floor.
HFO handling has no equivalent flammable-vapor exposure program to build. Cyclopentane's REACH classification and safety data sit on ECHA's substance information page, the reference EU-side buyers typically ask for alongside the supplier's own SDS.
Choosing the Right Route by Application
Appliance insulation panels (refrigerators, freezers) mostly convert to cyclopentane. The volumes justify the ATEX retrofit, and thinner panel sections tolerate the higher aged lambda without a major thickness penalty.
Cold-chain and construction insulation panels, where every millimeter of core thickness affects usable storage volume, more often go HFO. The lower aged lambda holds panel thickness closer to the old HCFC-141b spec.
Spray-applied rigid foam — now under the same July 2026 deadline as everything else — is one of the strongest cases for HFO. A spray rig is mobile and hard to retrofit with ATEX-rated equipment on site.
Crews also work in occupied or semi-occupied buildings, where a flammable overspray is a harder risk to manage than a fluorochemical one.
Industrial pipe-in-pipe and low-spec, non-insulation-critical parts are where water/CO2 blowing earns its keep. The equipment change is minimal, and the density/lambda penalty matters less when thermal performance isn't the primary spec driver.
Working With a Manufacturer-Direct Polyol Partner
We formulate and supply both cyclopentane and HFO rigid foam polyol systems out of our own plant, so a conversion project runs through one technical team instead of a trading intermediary relaying specs back and forth.
Requalification support covers side-by-side lab trials against your current HCFC-141b baseline, updated TDS and COA once the new blowing agent is confirmed, and index/catalyst tuning on your actual mixhead parameters rather than generic lab conditions.
Our MOQ structure is tiered across four volume bands, so a plant can pull a small qualification-batch drum before committing to full-container production once the new system passes your line trial.
FAQ
Q: Does the 2026 China ban affect foam already installed in buildings or appliances?
No. The ban covers new production and import of HCFC-141b for blowing agent use from the 2026 dates above; it doesn't require retrofitting foam already in service. Buyers should confirm the blowing agent named on the supplier's post-2026 TDS and COA going forward.
Q: Can cyclopentane and HFO be blended in the same system?
Yes — co-blowing cyclopentane with an HFO is common practice to cut the flammable load per unit volume while capturing part of the HFO's lambda advantage. The blend still needs its own qualification run against ASTM D1622 (density) and D1621 (compressive strength); it isn't a linear average of the two single-agent systems.
Q: Will switching off HCFC-141b change my fire rating?
Often yes. HCFC-141b's chlorine content contributed some passive flame suppression that hydrocarbon blowing agents don't provide, so flame retardant loading commonly needs to increase to hold the same UL94, GB 8624, or EN 13501-1 class. Retest — don't assume the old rating transfers.
Q: How long does formulation requalification actually take?
Lab-scale free-rise and box-foam trials run a few weeks; full-scale line trials and third-party fire/thermal testing (ASTM E84, ASTM C518, or ISO 8301) typically add another 6-10 weeks given normal lab booking lead times. Budget 2-4 months end to end.
Q: Is water/CO2 blowing viable for cold-chain panel insulation?
Generally not as a primary route. All-water systems run 15-25% higher core density and a meaningfully higher aged lambda than cyclopentane or HFO, which pushes panel thickness up for the same R-value — a real problem when internal storage volume is part of the spec.
