The truth about refrigerant GWP

Some common refrigerants have global warming potentials that are thousands of times worse than carbon dioxide but that knowledge is meaningless without context. The quantities used and their rate of release into the atmosphere must also be considered in order to understand the relative global warming impact of refrigerants.

Environmental assessment methods like BREEAM and LEED notably punish the use of refrigerants with a high global warming potential (GWP) so the natural assumption is that their effect must be a significant proportion of the overall impact caused by a building project.

Unfortunately most discussions about refrigerant GWP concentrates on the effect relative to carbon dioxide without any explanation of context, so let’s take a look at some numbers:

The following graph compares the global warming impact of the energy used by a 20,000m2 office building for one year with the impact of the refrigerant contained in the associated chillers shown if that refrigerant were to be released into the atmosphere:

With a traditional refrigerant like R134a, the impact is roughly 78% of the energy used in a year (or 3.6 times the energy used for heating and cooling in a year). That sounds like a significant effect but how much refrigerant actually leaks from a typical chiller over a year?

Leakage rates can be observed in various studies as falling between 2% and 20% per year, with more recent studies on newer units ranging between 3% and 10%. Some manufacturers will even claim monitored leakage rates for their units fall as low as 0.5% per year for their equipment.

That means that the impact of the refrigerant in the graph above is typically spread over 10 to 33 years.  The impact would therefore be between 2% and 8% of the buildings energy related warming impact.

Expressed as a percentage of the heating and cooling energy, this is 11% to 36%.

With a readily available low GWP refrigerant like R32 the figure would be only 1% to 4% of the impact of total building energy consumption.

It could be argued that if a decision is required between refrigerant choice and chiller/heat pump efficiency then it is only necessary to achieve approximately a 10% improvement in efficiency to justify using the more aggressive refrigerant.

Was it rational for environmental assessments to penalise high GWP refrigerants out of proportion to their effects?   

Without this artificial constraint, the industry may not have moved to produce the range of low GWP refrigerants that are now available and while the numbers are small in proportion to the impact of energy usage, they are still important in absolute terms.

There are also two important time related considerations for the future of refrigerant impact that must be considered:

1)     As the energy demand of buildings reduces, the proportionate effect of the refrigerant GWP increases. Plant size does not drop in linear proportion to the reduction in energy usage. A building with half the cooling energy usage over a year does not have a chiller that his half the size installed.

2)      As the carbon intensity of the electrical grid reduces, the GWP of refrigerants has a larger proportionate impact. For example when the grid carbon intensity approaches zero (which is set to occur by 2050), the GWP of the refrigerants approaches infinity as a percentage of the energy impact.

What about heating only heat pumps in a low energy building? Are they still better than gas boilers when the GWP is considered?

The leakage rate from domestic heat pumps in particular tends to be larger than commercial equipment and with the current carbon impact of the electrical grid in Ireland the effect of the refrigerant leakage reduces the environmental savings of using a heat pump by approximately 12% to 24% but the impact of using a heat pump is still notably lower than that of natural gas.

To conclude:

Modern refrigerants contribute to less than 4% of the global warming impact of buildings but this percentage will grow rapidly as the carbon intensity of electricity drops unless we move to lower GWP refrigerants.  

Luckily low GWP refrigerants and for many applications zero GWP refrigerants, are readily available with only a small price premium and will shortly become the default for all buildings.