The UK Net Zero Carbon Buildings Standard Guide: Part 10: Refrigerants

refrigerant banner

The pilot version of the UK Net Zero Carbon Buildings Standard (the Standard) was published in September 2024. The full document can be downloaded here. The Standard has been produced by a range of industry professional organisations including RIBA (architecture), IStructE (structural engineering), CIBSE (services engineers) and RICS (surveyors), along with a large team of other industry organisations and professionals. 

It aims to set out unambiguously, for a wide range of scenarios, the characteristics that buildings and building projects need to be aligned with the UK’s strategy to become net zero carbon by 2050. The Standard builds upon and supersedes previously published approaches such as the UKGBC Net Zero Carbon Building Framework, the RIBA Climate Challenge and the various LETI design guides.

Read more from our guide: 

Part 1: Key principles and overview 

Part 2: Embodied carbon

Part 3: Operational energy

Part 4: On-Site renewable generation

Part 5: Operational Water Use

Part 6: Fossil fuel free

Part 7: Electricity demand management 

Part 8: District heating and cooling networks

Part 9: Space heating and cooling

Part 10: Refrigerants 

Part 11: Carbon offsetting

Refrigerant requirements

The Standard defines limits for refrigerants, which are used in heat pumps for producing heating, cooling and hot water. These limits apply for all building types, including refrigeration systems but excluding refrigeration used as part of industrial processes. 

In order to meet the Standard, all building systems must use refrigerant with GWP 677kgCO2eq/kg or lower. This has been pegged to the GWP of R32, so if the GWP of R32 is deemed to have changed in future, this limit would change with it. The Standard requires the total building refrigerant charge to be declared, expressed in kgCO2eq. This will be a summation of all refrigerant-containing systems:

Equation 5: Refrigerant charge in the building 

Building refrigerant charge (kgCO2e) = RC1 x GWP1 + RC2 x GWP2 .... + RCn x GWPn

Where 

  • RCn = refrigerant charge (kg) in system n
  • GWPn = GWP of refrigerant (kgCO2e/kg) in system n

Any buildings which contain more than 3,000 kgCO2eq worth of refrigerant must comply with additional leakage reporting measures. This reporting is a set of maintenance standards and documentation over the life of the equipment - not simply installing a 'leak detection' system. Refrigerant quantity must be measured and reported during maintenance, top ups and end of life.

Comments on the proposals

Aspects we think work well

  • Strict leakage reporting – the impact of refrigerant leakage can be significant, but the magnitude of leakage is difficult to predict. TM65 outlines standard rates, but these are not necessarily representative of reality. In projects, it is not uncommon to witness full system leakage, which for systems which use higher GWP refrigerants would be catastrophic in terms of carbon emissions.
  • The reporting limit of 3,000 kgCO2eq seems appropriate, as it is likely to be required with any R32 (or similar) refrigerant systems, but less likely for R290 or CO2 systems. This feels like an appropriate way to allow R32, but incentivise the transition to natural refrigerants.

Aspects recommended be considered for further development

  • The Standard have set the GWP limit to correspond with an outdated value for R32, based on the Intergovernmental Panel for Climate Change (IPCC) Assessment Report 5. However, this does not align with the RICS Professional Standards version 2, which requires the latest Assessment Report GWP values to be used at any given moment in time. The current Assessment Report – AR6 – defines the GWP of R32 as 771kgCO2eq/kg
  • The standard describes additional reporting for refrigerant leakage, but does not provide a narrative on the use of leak detection systems.
  • The standard does not advise or comment on the use of Variable Refrigerant Flow (VRF) systems, which are the most impactful in terms of refrigerant leakage due to higher quantities of refrigerant and refrigerant being piped around a building necessitating low flammability and low toxicity. 

Wider industry considerations

  • There is strong industry lobbying for retaining R32. While it may be considered an acceptable choice in the short term, it should not stop manufacturers from developing natural refrigerant options.
  • European Union F-gas regulations propose a strict refrigerant phase down, banning the sale of new R32 heat pumps under 50 kW by 2027, and above 50 kW by 2030. While the UK is not necessarily committed to these standards, they are likely to heavily influence industry direction (many more heat pumps are sold in the EU than in the UK). By accepting the use of R32 in net zero carbon buildings, the UK will fall behind EU standards.
  • Ground Source Heat Pumps (GSHP) are much further behind in low GWP refrigerant development than Air Source Heat Pumps and change to this sector is required if they are to keep pace.
  • Many refrigerant-containing products are designed for indoor use, relying on R410a due to its non-flammable and non-toxic properties. Phasing out R410a based on its high GWP is necessary, and should be encouraged. 

Example from our projects

From our work on whole life carbon, we have found that the biggest refrigerant leakage impacts are from very high GWP refrigerants, such as R410a, and VRF systems. From initial studies, R290 systems have very low leakage impacts, but generally require larger units, hence have a higher upfront carbon. These trade-offs result in similar whole life carbon impacts for self-contained heat pumps which utilitse refrigerants with a GWP below that for R32. This offers a range of lower whole life carbon design options, which are in line with the UKNZCBS proposals.

All assessments of refrigerant impacts assume standard leakage rates, taken from TM65. Rates are higher for VRF systems, due to the higher refrigerant fill and more junctions within the pipework.

It is difficult to predict the realistic leakage of an installation. From project experience, it is not uncommon for a system to fail and release its entire refrigerant charge. This full refrigerant dump is always possible, and our view is that this scenario should be modelled.

In a recent project, we have investigated options for domestic heat pumps. Due to site restrictions, R290 is not currently feasible and so R32 is proposed. We undertook a high level Whole Life Carbon assessment for both units, with standard and ‘full system’ leakage assumptions, under a medium temperature operation. This demonstrated that, although leakage would increase the associated carbon emissions by ~5-20%, the overall impact of leakage on the Whole Life Carbon results was small.