Increase Building Deconstruction & Recycling
This solution involves reducing the use of high-global warming potential (GWP) refrigerants, instead deploying lower-GWP refrigerants. High-GWP (>800 on a 100-yr basis) fluorinated gases (F-gases) are currently used as refrigerants in refrigeration, air conditioning, and heat pump systems. Over the lifetime of this equipment, refrigerants escape into the atmosphere where they contribute to climate change.
Leaked lower-GWP refrigerant gases trap less heat in the atmosphere than do higher-GWP gases, so using lower-GWP gases reduces the climate impact of refrigerant use. In our analysis, this solution is only deployed as new equipment replaces decommissioned equipment because alternative refrigerants cannot typically be retrofitted into existing systems.
Refrigerants are chemicals that can absorb and release heat as they move between gaseous and liquid states under changing pressure. In this solution, we considered their use in six applications: residential, commercial, industrial, and transport refrigeration as well as stationary and mobile air conditioning. Heat pumps double as heating sources, though they are included here with air conditioning appliances. Refrigerants are released to the atmosphere during manufacturing, transport, installation, operation, repair, and disposal of refrigerants and equipment. Deploy Alternative Insulation Materials covers the use of refrigerant chemicals to produce foams.
Climate impacts of emissions of refrigerants can be reduced by:
This solution evaluated the use of lower-GWP refrigerants alone. Leak reduction and responsible disposal are covered in Improve Refrigerant Management. Lowering use of and demand for refrigerants – while outside the scope of these assessments – is the most effective way to reduce emissions.
Most refrigerants used in new equipment today are a group of F-gases called hydrofluorocarbons (HFCs) (Figure 1). HFCs are GHGs and are typically hundreds to thousands of times more potent than CO₂ (Smith et al., 2021). Since high-GWP refrigerants are usually short-lived climate pollutants, their negative climate impacts tend to be concentrated in the near term (Shah et al., 2015). High-GWP HFC production and consumption are being phased down under the Kigali Amendment to the Montreal Protocol, but existing stock and production remains high worldwide (Amendment to the Montreal Protocol on Substances That Deplete the Ozone Layer, 2016; United Nations Framework Convention on Climate Change [UNFCCC], 2023). Other types of refrigerants that deplete the ozone layer – including chlorofluorocarbons (CFCs) and hydrochlorofluorocarbons (HCFCs) — are also being phased out of new production and use globally (Montreal Protocol on Substances That Deplete the Ozone Layer, 1987; Figure 1).
Figure 1. Examples of common refrigerants and their climate and environmental impacts
High-GWP: red; Medium-GWP: yellow; Low-GWP: green
| Type | GWP (20-yr) |
GWP (100-yr) |
Lifetime (yr) |
Ozone Depleting? |
PFAS? | Safety Class* |
|
|---|---|---|---|---|---|---|---|
| R11 | CFC | 8,320 | 6,230 | 52 | Yes | A1 | |
| R12 | CFC | 12,700 | 12,500 | 102 | Yes | A1 | |
| R22 | HCFC | 5,690 | 1,960 | 11.9 | Yes | A1 | |
| R141b | HCFC | 2,710 | 860 | 9.4 | Yes | ||
| R125 | HFC | 6,740 | 3,740 | 30 | No | Yes | A1 |
| R134a | HFC | 4,140 | 1,530 | 14 | No | Yes | A1 |
| R143a | HFC | 7,840 | 5,810 | 51 | No | Yes | A2L |
| R404A | HFC blend | 7,208 | 4,728 | No | Yes | A1 | |
| R407C | HFC blend | 4,457 | 1,908 | No | Yes | A1 | |
| R410A | HFC blend | 4,715 | 2,256 | No | Yes | A1 | |
| R452A | HFC/HFO blend | 4,273 | 2,292 | No | Yes | A1 | |
| R32 | HFC | 2,690 | 771 | 5.4 | No | No | A2L |
| R452B | HFC/HFO blend | 2,275 | 779 | No | Yes | A2L | |
| R454A | HFC/HFO blend | 943 | 270 | No | Yes | A2L | |
| R513A | HFC/HFO blend | 1,823 | 673 | No | Yes | A1 | |
| R290 (Propane) | Natural | 0.072 | 0.02 | 0.036 | No | No | A3 |
| R600a (Isobutane) | Natural | < 1 | < 1 | 0.019 | No | No | A3 |
| R717 (Ammonia) | Natural | < 1 | < 1 | < 1 | No | No | B2L |
| R744 (CO₂) | Natural | 1 | 1 | No | No | A1 | |
| R1234yf | HFO | 1.81 | 0.501 | 0.033 | No | Yes | A2L |
| R1234ze(E) | HFO | 4.94 | 1.37 | 0.052 | No | Yes | A2L |
*Safety classes based on ASHRAE Standard 34:
A1: non-flammable, lower toxicity
A2L: lower flammability, lower toxicity
A3: higher flammability, lower toxicity
B2L: lower flammability, higher toxicity
Sources:
Baha, M., & Dupont, J.-L. (2023, September 15). Global warming potential (GWP) of HFC refrigerants. International Institute of Refrigeration.
Behringer, D., Heydel, F., Gschrey, B., Osterheld, S., Schwarz, W., Warncke, K., Freeling, F., Nödler, K., Henne, S., Reimann, S., Blepp, M., Jörß, W., Liu, R., Ludig, S., Rüdenauer, I., & Gartiser, S. (2021). Persistent degradation products of halogenated refrigerants and blowing agents in the environment: Type, environmental concentrations, and fate with particular regard to new halogenated substitutes with low global warming potential. Final report. Umweltbundesamt [German Environment Agency].
Burkholder, J. B., Hodnebrog, Ø., McDonald, B. C., Orkin, V., Papadimitriou, V. C., & Van Hoomissen, D. (2023). Annex: Summary of abundances, lifetimes, ODPs, REs, GWPs, and GTPs. Scientific Assessment of Ozone Depletion 2022.
Garry, M. (2021, June 23). Certain HFCs and HFOs are in PFAS group that five EU countries intend to restrict.
Smith, C., Nicholls, Z. R. J., Armour, K., Collins, W., Forster, P., Meinshausen, M., Palmer, M. D., & Watanabe, M. (2021). The Earth’s energy budget, climate feedbacks, and climate sensitivity supplementary material. In Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change.
Trevisan, T. (2023, July 3). Overview of PFAS refrigerants used in HVAC&R and relevance of refrigerants in the PFAS Restriction Intention. UN Montreal Protocol 45th OEWG, Bangkok.
United Nations Environment Programme. (2023). Refrigeration, Air Conditioning and Heat Pumps Technical Options Committee: 2022 assessment report.
United Nations Environment Programme & ASHRAE. (2025). Update on new refrigerants designations and safety classifications June 2025.
In this solution, production and consumption of high-GWP refrigerants (which we defined as GWP>800, 100-yr basis) are avoided by the use of lower-GWP refrigerants in new equipment. These alternative refrigerants can still leak to the atmosphere, but their heat-trapping effect is much lower. Some promising alternatives have low GWPs (<5, 100-yr basis), including some hydrofluoroolefins (HFOs) as well as natural refrigerants, which include CO₂, ammonia, propane, and isobutane. (Figure 1). However, the adoption of these low-GWP refrigerants comes with challenges, including flammability, cost, building codes, and technical limitations (see Risks and Take Action sections below).
Refrigerants with medium GWPs (<800, 100-yr basis; <2,700, 20-yr basis (Smith et al., 2021)) can also be near-term alternatives that increase adoption while providing a climate benefit. In our analysis, we separately considered medium-GWP alternatives in applications where low-GWP alternatives are less common (Figure 2).
Figure 2. Alternative refrigerants used to calculate the low-GWP and medium-GWP scenarios. The low-GWP scenario assumed equipment using high-GWP refrigerants is replaced at end-of-life with equipment using alternative refrigerants with GWP<5. The medium-GWP calculations assumed GWP<800 (100-yr basis) and GWP<2,700 (20-yr basis) alternatives in applications where low-GWP replacements are currently less common (commercial refrigeration, transport refrigeration, stationary air conditioning) and assumed low-GWP replacements for the remaining applications where they are more developed technologies (residential refrigeration, industrial refrigeration, mobile air conditioning). The alternative refrigerants in the table are used for effectiveness and/or cost calculations.
| Application |
Scenario 1: Low-GWP only (low GWP: < 5, 100-year basis) |
Scenario 2: Medium-GWP when low-GWP alternatives are less common, otherwise low-GWP (medium GWP: < 800, 100-year basis) |
|---|---|---|
| Residential refrigeration | Isobutane | |
| Commercial refrigeration | Propane, CO₂ | Medium-GWP HFC and HFO blends |
| Industrial refrigeration | Ammonia, CO₂, propane | |
| Transport refrigeration | Propane, propene, ammonia, CO₂, low-GWP HFOs |
Medium-GWP HFC and HFO blends |
| Mobile air conditioning | CO₂, low-GWP HFOs | |
| Stationary air conditioning | Propane, CO₂, ammonia, low-GWP HFOs |
Medium-GWP HFC and HFO blends |
Sources:
Purohit, P., & Höglund-Isaksson, L. (2017). Global emissions of fluorinated greenhouse gases 2005–2050 with abatement potentials and costs. Atmospheric Chemistry and Physics, 17(4), 2795–2816.
Sustainable Purchasing Leadership Council Climate Friendly Refrigerant Action Team. (2021). Recommendations for climate friendly refrigerant management and procurement.
United Nations Environment Programme. (2023). Refrigeration, Air Conditioning and Heat Pumps Technical Options Committee: 2022 assessment report.
United Nations Framework Convention on Climate Change. (2023). National inventory submissions, Annex 1 parties [Data set].
U.S. Environmental Protection Agency. (2011). Transitioning to low-GWP alternatives in transport refrigeration.
There is currently no single refrigerant that perfectly fits the climate, safety, and performance requirements for all applications. Instead, the optimal alternative refrigerant will vary depending on equipment type and location (United Nations Environment Programme [UNEP], 2023).
Generating electricity to run heating, ventilation, air conditioning, and refrigeration (HVAC&R) equipment also produces high levels of emissions (mostly CO₂ ) at power plants – more than twice the emissions from direct release of refrigerants (United Nations Development Programme [UNDP], 2022). Using alternative refrigerants can impact efficiency, changing these electricity-related emissions. However, indirect emissions are not quantified in this solution.
Air-Conditioning, Heating, and Refrigeration Institute. (n.d.-a). A2L refrigerant building code map - Canada. Link to source: https://www.ahrinet.org/a2l-refrigerant-building-code-map-canada
Air-Conditioning, Heating, and Refrigeration Institute. (n.d.-b). A2L refrigerant building code map - US. Link to source: https://www.ahrinet.org/a2l-refrigerant-building-code-map-us
Amendment to the Montreal Protocol on substances that deplete the ozone layer. (2016, October 15). Link to source: https://treaties.un.org/doc/Treaties/2016/10/20161015%2003-23%20PM/Ch_XXVII-2.f-English%20and%20French.pdf
Arp, H. P. H., Gredelj, A., Glüge, J., Scheringer, M., & Cousins, I. T. (2024). The global threat from the irreversible accumulation of trifluoroacetic acid (TFA). Environmental Science & Technology, 58(45), 19925–19935. Link to source: https://doi.org/10.1021/acs.est.4c06189
ASHRAE. (2009). ASHRAE position document on natural refrigerants. Link to source: https://www.epa.gov/sites/default/files/documents/ASHRAE_PD_Natural_Refrigerants_2011.pdf
Babiker, M., Berndes, G., Blok, K., Cohen, B., Cowie, A., Geden, O., Ginzburg, V., Leip, A., Smith, P., Sugiyama, M., & Yamba, F. (2022). Cross-sectoral perspectives. In P. R. Shukla, J. Skea, R. Slade, A. Al Khourdajie, R. van Diemen, D. McCollum, M. Pathak, S. Some, P. Vyas, R. Fradera, M. Belkacemi, A. Hasija, G. Lisboa, S. Luz, & J. Malley (Eds.), Climate change 2022: Mitigation of climate change. Contribution of working group III to the sixth assessment report of the intergovernmental panel on climate change (pp. 1245–1354). Cambridge University Press. Link to source: https://doi.org/10.1017/9781009157926.014
Baha, M., & Dupont, J.-L. (2023, September 15). Global warming potential (GWP) of HFC refrigerants. International Institute of Refrigeration. Link to source: https://iifiir.org/en/encyclopedia-of-refrigeration/global-warming-potential-gwp-of-hfc-refrigerants
Behringer, D., Heydel, F., Gschrey, B., Osterheld, S., Schwarz, W., Warncke, K., Freeling, F., Nödler, K., Henne, S., Reimann, S., Blepp, M., Jörß, W., Liu, R., Ludig, S., Rüdenauer, I., & Gartiser, S. (2021). Persistent degradation products of halogenated refrigerants and blowing agents in the environment: Type, environmental concentrations, and fate with particular regard to new halogenated substitutes with low global warming potential. Final report. Umweltbundesamt [German Environment Agency]. Link to source: https://www.umweltbundesamt.de/sites/default/files/medien/5750/publikationen/2021-05-06_texte_73-2021_persistent_degradation_products.pdf
Blumberg, K., Isenstadt, Taddonio, K. N., Andersen, S. O., & Sherman, N. J. (2019). Mobile air conditioning: The life-cycle costs and greenhouse-gas benefits of switching to alternative refrigerants and improving system efficiencies. Link to source: https://theicct.org/wp-content/uploads/2021/06/ICCT_mobile-air-cond_CBE_201903.pdf
Bolaji, B. O., & Huan, Z. (2013). Ozone depletion and global warming: Case for the use of natural refrigerant – a review. Renewable and Sustainable Energy Reviews, 18, 49–54. Link to source: https://doi.org/10.1016/j.rser.2012.10.008
Booten, C., Nicholson, S., Mann, M., & Abdelaziz, O. (2020). Refrigerants: Market trends and supply chain assessment. Link to source: https://www.nrel.gov/docs/fy20osti/70207.pdf
Burkholder, J. B., Hodnebrog, Ø., McDonald, B. C., Orkin, V., Papadimitriou, V. C., & Van Hoomissen, D. (2023). Annex: Summary of abundances, lifetimes, ODPs, REs, GWPs, and GTPs. Scientific Assessment of Ozone Depletion 2022. Link to source: https://csl.noaa.gov/assessments/ozone/2022/downloads/Annex_2022OzoneAssessment.pdf
California Public Utilities Commission. (2022). Refrigerant avoided cost calculator [Data set]. Link to source: https://www.cpuc.ca.gov/dercosteffectiveness
Campbell, I., Kalanki, A., & Sachar, S. (2018). Solving the global cooling challenge: How to counter the climate threat from room air conditioners. Rocky Mountain Institute. Link to source: https://rmi.org/wp-content/uploads/2018/11/Global_Cooling_Challenge_Report_2018.pdf
Chele, F. S., Salvador, C., Stevenson, L., Dolislager, F., Armstrong, A., Power, S., Mathews, T., & Yana Motta, S. F. (2024). Critical literature review of low global warming potential (GWP) refrigerants and their environmental impact. Oak Ridge National Laboratory. Link to source: https://www.osti.gov/servlets/purl/2528023
CLASP. (2025, December). Mepsy: The appliance & equipment climate impact calculator (Version 1.15.0). Link to source: https://clasp.shinyapps.io/mepsy
CLASP & ATMOsphere. (2022). Global refrigerant impact from 6 appliances in 22 countries. Link to source: https://www.clasp.ngo/wp-content/uploads/2022/05/Refrigerant-Modeling-Interim-Report-APR142022.pdf
Climate and Ozone Protection Alliance. (2025). Global banks of ozone depleting substances (ODS) and hydrofluorocarbons (HFCs). Link to source: https://www.copalliance.org/imglib/publications/2025-04_COPA_Global%20Banks%20ODS%20HFC_update.pdf
Colbourne, D., Croiset, I., Ederberg, L., Heubes, J., Martin, M., Narayan, C., Oppelt, D., Papst, I., Usinger, J., & Gschrey, B. (2013). NAMAs in the refrigeration, air conditioning and foam sectors: A technical handbook. Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH [German Agency for International Cooperation] . Link to source: https://transparency-partnership.net/sites/default/files/e-bruochure-20131015-neu.pdf
Copernicus Climate Change Service. (2023). ERA5 hourly data on single levels from 1940 to present [Data set]. Copernicus Climate Change Service (C3S) Climate Data Store (CDS). Retrieved January 13, 2026 from Link to source: https://doi.org/10.24381/cds.adbb2d47
Crippa, M., Guizzardi, D., Pagani, F., Banja, M., Muntean, M., Schaaf, E., Quadrelli, R., Risquez Martin, A., Taghavi-Moharamli, P., Köykkä, J., Grassi, G., Rossi, S., Melo, J., Oom, D., Branco, A., Suárez-Moreno, M., Sedano, F., San-Miguel, J., Manca, G., … Pekar, F. (2025). GHG emissions of all world countries - 2025 Report (JRC143227). Publications Office of the European Union. Link to source: https://data.europa.eu/doi/10.2760/9816914
Denzinger, P. (2023, December 9). Context and global mitigation potential of “green” ACs. COP28 UAE. Link to source: https://www.green-cooling-initiative.org/fileadmin/user_upload/Final_02_COP28_Side_Event_Green_ACs_Final_Final.pdf
Dimitrakopoulou, M.-E., Karvounis, M., Marinos, G., Theodorakopoulou, Z., Aloizou, E., Petsangourakis, G., Papakonstantinou, M., & Stoitsis, G. (2024). Comprehensive analysis of PFAS presence from environment to plate. npj Science of Food, 8, 80. Link to source: https://doi.org/10.1038/s41538-024-00319-1
Dong, Y., Coleman, M., & Miller, S. A. (2021). Greenhouse gas emissions from air conditioning and refrigeration service expansion in developing countries. Annual Review of Environment and Resources, 46, 59–83. Link to source: https://doi.org/10.1146/annurev-environ-012220-034103
Dräger. (n.d.). Cooling with ammonia: What you should keep in mind. Link to source: https://www.draeger.com/Content/Documents/Content/ammoniak-fa-pdf-8110-en.pdf
Dreyfus, G., Borgford-Parnell, N., Christensen, J., Fahey, D. W., Motherway, B., Peters, T., Picolotti, R., Shah, N., & Xu, Y. (2020). Assessment of climate and development benefits of efficient and climate-friendly cooling. Link to source: https://www.ccacoalition.org/resources/assessment-climate-and-development-benefits-efficient-and-climate-friendly-cooling
European Chemicals Agency. (2023). Annex XV restriction report: Per- and polyfluoroalkyl substances (PFASs). Link to source: https://echa.europa.eu/documents/10162/f605d4b5-7c17-7414-8823-b49b9fd43aea
European Environmental Bureau. (2025). Universal PFAS restriction under REACH: Briefing on fluorinated gases in the universal PFAS restriction—The F-lephant in the room. Link to source: https://eeb.org/wp-content/uploads/2025/02/EEB_EURENI_F-gas_Policy-Brief.pdf
Fabris, F., Fabrizio, M., Marinetti, S., Rossetti, A., & Minetto, S. (2024). Evaluation of the carbon footprint of HFC and natural refrigerant transport refrigeration units from a life-cycle perspective. International Journal of Refrigeration, 159, 17–27. Link to source: https://doi.org/10.1016/j.ijrefrig.2023.12.018
Fenton, S. E., Ducatman, A., Boobis, A., DeWitt, J. C., Lau, C., Ng, C., Smith, J. S., & Roberts, S. M. (2021). Per- and polyfluoroalkyl substance toxicity and human health review: Current state of knowledge and strategies for informing future research. Environmental Toxicology and Chemistry, 40(3), 606–630. Link to source: https://doi.org/10.1002/etc.4890
Food and Agriculture Organization of the United Nations. (2019). The state of food and agriculture 2019: Moving forward on food loss and waste reduction. Link to source: https://openknowledge.fao.org/server/api/core/bitstreams/11f9288f-dc78-4171-8d02-92235b8d7dc7/content
Garavagno, M. d. l. A., Holland, R., Khan, M. A. H., Orr-Ewing, A. J., & Shallcross, D. E. (2024). Trifluoroacetic acid: Toxicity, sources, sinks and future prospects. Sustainability, 16(6), Article 2382. Link to source: https://doi.org/10.3390/su16062382
Garry, M. (2021, June 23). Certain HFCs and HFOs are in PFAS group that five EU countries intend to restrict. Link to source: https://naturalrefrigerants.com/certain-hfcs-and-hfos-are-in-pfas-group-that-five-eu-countries-intend-to-restrict/
Goetzler, W., Guernsey, M., Young, J., Fuhrman, J., & Abdelaziz, O. (2016). The future of air conditioning for buildings. U.S. Department of Energy. Office of Energy Efficiency and Renewable Energy. Building Technologies Office. Link to source: https://www.energy.gov/sites/prod/files/2016/07/f33/The%20Future%20of%20AC%20Report%20-%20Full%20Report_0.pdf
Gradient. (2015). Risk assessment of refrigeration systems using A2L flammable refrigerants. Link to source: https://www.ahrinet.org/system/files/2023-08/AHRI-8009_Final_Report.pdf
Green Cooling Initiative. (n.d.). Global greenhouse gas emissions from the RAC sector. Retrieved April 15, 2025, from Link to source: https://www.green-cooling-initiative.org/country-data
Hanson, M. L., Madronich, S., Solomon, K., Sulbaek Andersen, M. P., & Wallington, T. J. (2024). Trifluoroacetic acid in the environment: Consensus, gaps, and next steps. Environmental Toxicology and Chemistry, 43, 2091–2093. Link to source: https://doi.org/10.1002/etc.5963
Hayes, C., Stausholm, T., Ilana, K., & Devin, Y. (2023). Natural refrigerants: State of the industry. ATMOsphere. Link to source: https://atmosphere.cool/marketreport-2022/
Hersbach, H., Bell, B., Berrisford, P., Hirahara, S., Horányi, A., Muñoz‐Sabater, J., Nicolas, J., Peubey, C., Radu, R., Schepers, D., Simmins, A., Soci, C., Abdalla, S., Abellan, X., Balsamo, G., Bechtold, P., Biavati, G., Bidlot, J., Bonavita, M., … Thépaut, J. N. (2020). The ERA5 global reanalysis. Quarterly Journal of the Royal Meteorological Society, 146(730), 1999–2049. Link to source: https://doi.org/10.1002/qj.3803
Heubes, J., Martin, M., & Oppelt, D. (2012). Refrigeration, air conditioning and foam blowing sectors technology roadmap. GIZ Proklima. Link to source: https://unfccc.int/ttclear/misc_/StaticFiles/gnwoerk_static/TEM_tec_cfi_rm/993ecdfa67144e68b88b4735ea50fcf0/647faaa714484a2983fe6851111ab9aa.pdf
Höglund-Isaksson, L., Purohit, P., Amann, M., Bertok, I., Rafaj, P., Schöpp, W., & Borken-Kleefeld, J. (2017). Cost estimates of the Kigali Amendment to phase-down hydrofluorocarbons. Environmental Science & Policy, 75, 138–147. Link to source: https://doi.org/10.1016/j.envsci.2017.05.006
Holland, R., Khan, M. A. H., Driscoll, I., Chhantyal-Pun, R., Derwent, R. G., Taatjes, C. A., Orr-Ewing, A. J., Percival, C. J., & Shallcross, D. E. (2021). Investigation of the production of trifluoroacetic acid from two halocarbons, HFC-134a and HFO-1234yf and its fates using a global three-dimensional chemical transport model. ACS Earth and Space Chemistry, 5(4), 849–857. Link to source: https://doi.org/10.1021/acsearthspacechem.0c00355
Imamura, T., Kamiya, K., & Sugawa, O. (2015). Ignition hazard evaluation on A2L refrigerants in situations of service and maintenance. Journal of Loss Prevention in the Process Industries, 36, 553–561. Link to source: https://doi.org/10.1016/j.jlp.2014.12.018
Inforum, JMS Consulting, The Alliance for Responsible Atmospheric Policy, & Air-Conditioning, Heating, and Refrigeration Institute. (2019, December 12). Economic & consumer impacts of HFC phasedown. Link to source: https://www.congress.gov/116/meeting/house/110388/documents/HHRG-116-IF18-20200114-SD003.pdf
International Energy Agency. (2023, July 12). Space cooling. Link to source: https://www.iea.org/energy-system/buildings/space-cooling
Intergovernmental Panel on Climate Change. (2023). Climate change 2022: Mitigation of climate change. Working Group III Contribution to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press. Link to source: https://doi.org/10.1017/9781009157926
JMS Consulting & Inforum. (2018). Consumer cost impacts of U.S. ratification of the Kigali Amendment. Link to source: https://www.alliancepolicy.org/site/usermedia/application/10/Consumer_Costs_Final_InforumJMS_20181109.pdf
Kim, B., Lee, S. H., Lee, D., & Kim, Y. (2020). Performance comparison of heat pumps using low global warming potential refrigerants with optimized heat exchanger designs. Applied Thermal Engineering, 171, Article 114990. Link to source: https://doi.org/10.1016/j.applthermaleng.2020.114990
Montreal protocol on substances that deplete the ozone layer. (1987, September 16). Link to source: https://treaties.un.org/doc/publication/unts/volume%201522/volume-1522-i-26369-english.pdf
The North American Sustainable Refrigeration Council. (n.d.). HFC policies & refrigerant regulations by state. Link to source: https://nasrc.org/hfc-policy-tracker/
Ozone Secretariat. (2017, February 17). Frequently asked questions relating to the Kigali Amendment to the Montreal Protocol. United Nations, United Nations Environment Programme. Link to source: https://ozone.unep.org/sites/default/files/FAQs_Kigali_Amendment.pdf
Ozone Secretariat. (2020). Treaties. United Nations, United Nations Environment Programme. Link to source: https://ozone.unep.org/classification-parties
Ozone Secretariat. (n.d.). Country data. United Nations, United Nations Environment Programme. Link to source: https://ozone.unep.org/countries/data
Petri, Y., & Caldeira, K. (2015). Impacts of global warming on residential heating and cooling degree-days in the United States. Scientific Reports, 5(1), Article 12427. Link to source: https://doi.org/10.1038/srep12427
Purohit, P., & Höglund-Isaksson, L. (2017). Global emissions of fluorinated greenhouse gases 2005–2050 with abatement potentials and costs. Atmospheric Chemistry and Physics, 17(4), 2795–2816. Link to source: https://doi.org/10.5194/acp-17-2795-2017
Salvador, C. M., Chele, F. S., Stevenson, L., Dolislager, F., Armstrong, A., Mathews, T., & Yana Motta, S. (2024). Atmospheric transformation of refrigerants: Current research developments and knowledge gaps. International Refrigeration and Air Conditioning Conference, USA, Paper 2671. Link to source: https://docs.lib.purdue.edu/iracc/2671
Secop. (2018). Practical application of refrigerants R600a and R290 in small hermetic systems. Link to source: https://www.secop.com/fileadmin/user_upload/technical-literature/guidelines/application_guideline_r600a_r290_02-2018_desa610a202.pdf
Shah, N., Khanna, N., Karali, N., Park, W. Y., Qu, Y., & Zhou, N. (2017). Opportunities for simultaneous efficiency improvement and refrigerant transition in air conditioning. Lawrence Berkeley National Laboratory. Link to source: https://cooling.lbl.gov/publications/opportunities-simultaneous-efficiency
Shah, N., Wei, M., Letschert, V., & Phadke, A. (2015). Benefits of leapfrogging to superefficiency and low global warming potential refrigerants in room air conditioning. Lawrence Berkeley National Laboratory. Link to source: https://www.osti.gov/servlets/purl/1235571
Shah, N., Wei, M., Letschert, V., & Phadke, A. (2019). Benefits of energy efficient and low-global warming potential refrigerant cooling equipment. Lawrence Berkeley National Laboratory. Link to source: https://cooling.lbl.gov/publications/benefits-energy-efficient-and-low
Sherry, D., Nolan, M., Seidel, S., & Andersen, S. O. (2017). HFO-1234yf: An examination of projected long-term costs of production. Nolan Sherry & Associates, Center for Climate and Energy Solutions, Institute for Governance and Sustainable Development. Link to source: https://www.c2es.org/wp-content/uploads/2017/04/hfo-1234yf-examination-projected-long-term-costs-production.pdf
Smith, C., Nicholls, Z. R. J., Armour, K., Collins, W., Forster, P., Meinshausen, M., Palmer, M. D., & Watanabe, M. (2021). The Earth’s energy budget, climate feedbacks, and climate sensitivity supplementary material. In Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. Link to source: https://www.ipcc.ch/report/ar6/wg1/downloads/report/IPCC_AR6_WGI_Chapter07_SM.pdf
Sustainable Purchasing Leadership Council Climate Friendly Refrigerant Action Team. (2021). Recommendations for climate friendly refrigerant management and procurement. Link to source: https://www.igsd.org/publications/recommendations-for-climate-friendly-refrigerant-management-and-procurement/
Trevisan, T. (2023, July 3). Overview of PFAS refrigerants used in HVAC&R and relevance of refrigerants in the PFAS Restriction Intention. UN Montreal Protocol 45th OEWG, Bangkok. Link to source: https://ozone.unep.org/system/files/documents/OEWG45_ATMO_sidevent.pdf
United Nations Development Programme. (2022). Guidance note: Assessing greenhouse gas emissions from refrigerants use in UNDP operations. Link to source: https://www.undp.org/sites/g/files/zskgke326/files/2022-07/Refrigerants%20methodology%20version%20July%202022.pdf
United Nations Environment Programme. (2022). Medical and Chemical Technical Options Committee: 2022 assessment report. Link to source: https://ozone.unep.org/system/files/documents/MCTOC-Assessment-Report-2022.pdf
United Nations Environment Programme. (2023). Refrigeration, Air Conditioning and Heat Pumps Technical Options Committee: 2022 assessment report. Link to source: https://ozone.unep.org/system/files/documents/RTOC-assessment%20-report-2022.pdf
United Nations Environment Programme & ASHRAE. (2025). Update on new refrigerants designations and safety classifications June 2025. Link to source: https://www.ashrae.org/file%20library/professional%20development/ashrae-unep/unep---ashrae-factsheet--english.pdf
United Nations Framework Convention on Climate Change. (2023). National inventory submissions, Annex 1 parties [Data set]. Link to source: https://unfccc.int/ghg-inventories-annex-i-parties/2023
U.S. Environmental Protection Agency. (2011). Transitioning to low-GWP alternatives in transport refrigeration. Link to source: https://www.epa.gov/sites/default/files/2015-07/documents/transitioning_to_low-gwp_alternatives_in_transport_refrigeration.pdf
U.S. Environmental Protection Agency. (2025). Frequent questions on the phasedown of hydrofluorocarbons. Link to source: https://www.epa.gov/climate-hfcs-reduction/frequent-questions-phasedown-hydrofluorocarbons
Velders, G. J. M., Daniel, J. S., Montzka, S. A., Vimont, I., Rigby, M., Krummel, P. B., Muhle, J., O’Doherty, S., Prinn, R. G., Weiss, R. F., & Young, D. (2022). Projections of hydrofluorocarbon (HFC) emissions and the resulting global warming based on recent trends in observed abundances and current policies. Atmospheric Chemistry and Physics, 22(9), 6087–6101. Link to source: https://doi.org/10.5194/acp-22-6087-2022
Velders, G. J. M., Fahey, D. W., Daniel, J. S., Andersen, S. O., & McFarland, M. (2015). Future atmospheric abundances and climate forcings from scenarios of global and regional hydrofluorocarbon (HFC) emissions. Atmospheric Environment, 123, 200–209. Link to source: https://doi.org/10.1016/j.atmosenv.2015.10.071
World Meteorological Organization. (2018). Executive summary: Scientific assessment of ozone depletion: 2018 (Report No. 58). Link to source: https://ozone.unep.org/sites/default/files/2019-04/SAP-2018-Assessment-report-ES-rev%20%281%29.pdf
Zaelke, D., & Borgford-Parnell, N. (2015). The importance of phasing down hydrofluorocarbons and other short-lived climate pollutants. Journal of Environmental Studies and Sciences, 5(2), 169–175. Link to source: https://doi.org/10.1007/s13412-014-0215-7
Zanchi, V., Boban, L., & Soldo, V. (2019). Refrigerant options in the near future. Journal of Sustainable Development of Energy, Water and Environment Systems, 7(2), 293–304. Link to source: https://doi.org/10.13044/j.sdewes.d6.0250
Sarah Gleeson, Ph.D.
Ruthie Burrows, Ph.D.
James Gerber, Ph.D.
Daniel Jasper
Alex Sweeney
Aiyana Bodi
James Gerber, Ph.D.
Hannah Henkin
Heather McDiarmid, Ph.D.
Ted Otte
Amanda D. Smith, Ph.D.
For every kt high-GWP refrigerant phased out in favor of low-GWP refrigerant, approximately 460,000 t CO₂‑eq/yr of F-gas emissions will be mitigated on a 100-yr basis (Table 1). If medium-GWP refrigerants are instead adopted in certain applications (Figure 2), the effectiveness decreases to 400,000 t CO₂‑eq (100-yr)/kt high-GWP refrigerant phased out/yr (Table 1). Effectiveness is based on average GWP of the high-, low-, and medium-GWP refrigerants; the difference in refrigerant charge; and the expected percent released to the atmosphere.
Since F-gases are short-lived climate pollutants, the effectiveness of this solution on a 20-yr basis is higher than on a 100-yr basis. Switching to low-GWP refrigerants saves 860,000 t CO₂‑eq /kt high-GWP refrigerant phased out/yr on a 20-yr basis. Medium-GWP refrigerants in certain applications reduces the effectiveness to 700,000 t CO₂‑eq (20-yr)/kt high-GWP refrigerant phased out/yr.
Using low-GWP refrigerants mitigates almost all CO₂‑eq emissions from direct release of high-GWP refrigerants. Medium-GWP refrigerants potentially offer a faster path to adoption in certain applications, but yield a smaller reduction in CO₂‑eq emissions. Switching to the lowest possible GWP refrigerant appropriate for a given application will have the highest effectiveness at cutting emissions.
Table 1. Effectiveness at reducing emissions using low-GWP refrigerants only or medium-GWP refrigerants in some applications and low-GWP alternatives otherwise
Unit: t CO₂‑eq /kt high-GWP refrigerant phased out/yr, 100-yr basis
| Average – low GWP only | 460000 |
| Average – medium & low GWP | 400000 |
We estimated the cost of purchasing and using low-GWP alternative refrigerants and equipment by taking a weighted average across all application types, averaging to US$23 million/kt high-GWP refrigerant phased out/yr. A kt of refrigerant goes a long way; a typical residential air conditioner requires only 0.6–3 kg refrigerant, depending on the country and refrigerant type (CLASP & ATMOsphere, 2022). On average across all applications, the emissions abatement cost for this solution is only US$50/t CO₂‑eq on a 100-yr basis (Table 2), or US$27/t CO₂‑eq on a 20-yr basis.
We separately evaluated the net costs of using medium-GWP refrigerants in some applications (Figure 2). Using medium-GWP refrigerants brought average costs down to US$9.4 million/kt high-GWP refrigerant phased out/yr. The emissions abatement cost is US$24/t CO₂‑eq (100-yr basis) or US$13/t CO₂‑eq (20-yr basis).
We calculated cost using values of initial cost and annual operation and maintenance costs from Purohit and Höglund-Isaksson (2017). The overall net cost is a weighted average of the average net costs of switching to alternative refrigerants for each of the six refrigerant applications (Figure 2). Costs are likely to change as the HFC phase-down continues under the Kigali Amendment. We did not evaluate external costs such as those to manufacturers.
Although our calculated costs are averages, costs varied widely depending on the specific equipment, refrigerant type, and geographic location. Using ammonia in industrial refrigeration yields net savings of US$24 million/kt high-GWP refrigerant/yr. Low-GWP alternative refrigerants for transport refrigeration lead to cost savings over high- or medium-GWP refrigerants, as do hydrocarbons in residential and commercial air conditioning.
We did not consider energy cost differences due to changes in efficiency. Since electricity costs are the majority of the life-cycle costs for certain equipment, these changes in energy costs may be significant (Goetzler et al., 2016).
Table 2. Cost per unit of climate impact for low-GWP refrigerants.
Unit: 2023 US$/t CO₂‑eq , 100-yr basis
| Average | 50.00 |
We did not find a learning rate for this solution, although there is evidence that costs of equipment and refrigerant decrease as more alternative refrigerants are deployed. Zanchi et al. (2019) claim that after regulations limiting emissions from F-gases and capping allowable refrigerant GWP were enacted in Europe, component prices for natural refrigerant equipment – particularly in commercial refrigeration – became comparable with lower HFC unit prices. Equipment prices have trended downwards through other similar technological transitions in the past (JMS Consulting & Inforum, 2018).
The cost of refrigerants can change with adoption as well as the cost of equipment. Natural refrigerants tend to be inexpensive, but cost premiums for expensive HFO refrigerants could drop by more than 75% as production volumes increase (Booten et al., 2020). Certain expensive-to-produce alternative refrigerants like HFO-1234yf have limited information about possible future price reductions, but other refrigerant transitions have indicated that prices should decrease due to increased production experience, capacity, and number of producers – especially as patents expire (Sherry et al., 2017).
Speed of action refers to how quickly a climate solution physically affects the atmosphere after it is deployed. This is different from speed of deployment, which is the pace at which solutions are adopted.
At Project Drawdown, we define the speed of action for each climate solution as emergency brake, gradual, or delayed.
Deploy Alternative Refrigerants is an EMERGENCY BRAKE climate solution. It has the potential to deliver a more rapid impact than gradual and delayed solutions. Because emergency brake solutions can deliver their climate benefits quickly, they can help accelerate our efforts to address dangerous levels of climate change. For this reason, they are a high priority.
Permanence
There is a low risk of the emissions reductions for this solution being reversed. Each kt high-GWP refrigerant phased out for a lower-GWP alternative reduces the emissions from refrigerant release during manufacturing, transport, installation, operation, repair, and disposal of equipment.
Additionality
This solution is additional when alternative refrigerant is used in applications that would have used HFCs or other high-GWP refrigerants in recent history. HFCs are not the baseline refrigerant in every scenario: hydrocarbons, for example, have been widely used in residential refrigeration and ammonia in industrial refrigeration for many years.
In our analysis, we considered any path to adoption of alternative refrigerants to be part of its effectiveness at reducing GHG emissions. For example, we considered all HFC reductions mandated by policy to be considered additional over baseline HFC usage. However, some GHG accounting or crediting organizations would consider this regulatory additionality; the only emissions reductions that count as additional would be those not mandated by international, regional, and application-specific policy limits.
We estimated that 440 kt high-GWP refrigerants already have been phased out in favor of low-GWP alternative refrigerants worldwide (Table 3). For adoption, we did not differentiate between low- and medium-GWP alternative refrigerants due to insufficient data.
There are limited recent and global data available to quantify the adoption of alternative refrigerants. For this reason, our approach to quantifying adoption is a simplified approximation. We used projected 2022 HFC emissions from Velders et al. (2015) as our baseline. These projections were made before any Kigali Amendment phase-down began, and we assumed they represent a reasonable 2022 emissions picture in the absence of policy-regulated HFC reductions.
To calculate current adoption, we analyzed a Velders et al. (2022) model of 2022 HFC emissions accounting for current policies. Projected 2022 emissions in the current model were 6.4% lower than the 2015-projected baseline, which we assumed to be proportional to the amount of high-GWP HFC phased out and replaced with low-GWP alternatives. We estimated current adoption by applying this assumption to an estimated 6,480 kt bank of existing refrigerants (Climate and Ozone Protection Alliance, 2025). That bank includes all HFC and ozone-depleting refrigerants in new, in-use, and end-of-life equipment, and represents the potential refrigerant that could be replaced by alternative refrigerants. Since some alternative refrigerants were adopted before our 2015 baseline, the current adoption value is likely an underestimate.
Some applications are known to have higher levels of current adoption than others. For example, 800 million domestic refrigerators are estimated to use isobutane refrigerant globally, and most of the market for commercial supermarket plug-in cases in Europe, the United States, and Japan use hydrocarbons such as propane (Hayes et al., 2023; UNEP, 2023).
Table 3. Current (2022 modeled) adoption level of low-GWP alternative refrigerants relative to 2015 baseline levels.
Unit: kt high-GWP refrigerant phased out
| Estimate | 440 |
We estimated that 77 kt high-GWP refrigerants are phased out for alternative low-GWP refrigerants each year (Table 4). Using the same method as current adoption, we compared baseline and policy-adjusted projections of HFC emissions from Velders et al. (2015, 2022) for 2019–2022. The difference between the projections increased by a median 1.2% year-over-year.
We applied this percent change directly to the 2022 HFC refrigerant bank estimate to determine the tonnage of high-GWP refrigerant that will be phased out as new equipment replaces decommissioned stock. We assumed the replacements all use low-GWP refrigerants.
Although more HFC is being phased out each year, the bank and associated emissions of HFCs are also growing as refrigeration and cooling equipment are more heavily used globally. Alternative refrigerant adoption will need to outpace market growth before net emissions reductions occur. The adoption trend is likely higher today than what is reflected by the data used in our calculations (prior to 2023), since 2024 was a Kigali-mandated increase in HFC phase-down for certain countries. We expect adoption trend to continue to increase as HFC restrictions tighten further in the future.
Table 4. 2019–2022 adoption trend of low-GWP alternative refrigerants.
Unit: kt high-GWP refrigerant phased out/yr
| Estimate | 77 |
The adoption ceiling for this solution is phasing out all high-GWP refrigerants, or 6,900 kt globally (Table 5). This value represents the entire current bank of HFCs and ozone-depleting refrigerants added to the current adoption of low-GWP refrigerants (Climate and Ozone Protection Alliance, 2025).
This quantity assumes no increase in the total refrigerant bank above 2022 levels, while in reality the bank is projected to increase substantially as demand for cooling and refrigeration grows worldwide (International Energy Agency [IEA], 2023). Consumption of refrigerants in stationary air conditioning applications alone is projected to increase 3.5-fold between 2020–2050 (Denzinger, 2023). Additionally, new equipment that uses refrigerants (such as heat pump water heaters) is expected to replace non-refrigerant equipment, adding to future refrigerant demand. However, projecting future refrigerant demand was not part of this assessment.
We assumed that in all future cases, high-GWP refrigerants can be phased out for low-GWP alternatives. While ambitious, this ceiling is possible across all applications as new refrigerants, blends, and equipment are developed and commercialized. Since we considered implementation in new equipment, it comes with an adoption delay as existing equipment with high-GWP refrigerants finish their lifespans, which can last 10–20 years (California Public Utilities Commission, 2022; CLASP & ATMOsphere, 2022).
Table 5. Adoption ceiling for low-GWP refrigerants.
Unit: kt high-GWP refrigerant phased out
| Estimate | 6,900 |
The achievable adoption range is clearly laid out by the Kigali Amendment schedule for reduction in HFC consumption and production. The Achievable – Low adoption assumes that worldwide, all countries meet the Kigali phase-down schedule and collectively reach 80% reduction from baseline emissions by 2045. Under the Kigali Amendment, all participating countries are expected to meet at least this standard by this date. It is achievable that this adoption level could be reached collectively across all nations (including higher-adopting countries and non-Kigali signatories). This comes to 5,500 kt reduction in high-GWP refrigerants, calculated as 80% of the sum of net bank and current adoption (Table 6).
Achievable – High assumes that all countries average the highest Kigali-mandated HFC reduction levels for any country (85% reduction from baseline), which comes to 5,900 kt high-GWP refrigerant phased out when applied to our adoption ceiling. If countries continue to follow the Kigali Amendment phase-down schedule, most production and use of HFCs will be eliminated over the coming decades. Other high-GWP ozone-depleting refrigerants are mostly phased out of new production under the Montreal Protocol, although large quantities still exist in refrigerant banks (Montreal Protocol on Substances That Deplete the Ozone Layer, 1987).
Our achievable adoption values do not account for growth in the refrigerant bank over 2022 levels. Although refrigerant use is expected to grow substantially in the coming decades (IEA, 2023), we did not project future demand as part of our assessment. If HFC phaseout does not outpace refrigerant demand growth, emissions can increase despite more widespread adoption of this solution. Lowering the demand for refrigerant while ensuring that all people have access to refrigeration, heating, and cooling will be challenging.
Table 6. Range of achievable adoption levels for low-GWP refrigerants.
Unit: kt high-GWP refrigerant phased out
| Current adoption | 440 |
| Achievable – low | 5500 |
| Achievable – high | 5900 |
| Adoption ceiling | 6900 |
This solution has high potential climate impact due to both the quantity and high GWP of many current refrigerants. High-GWP refrigerant already phased out for low-GWP alternatives has an estimated current climate impact of 0.20 Gt CO₂‑eq/yr on a 100-yr basis (Table 7). If the Kigali Amendment HFC phasedown schedule is followed globally, we expect the achievable-adoption climate impact to be 2.5–2.7 Gt CO₂‑eq (100-yr)/yr. Reaching the adoption ceiling could potentially mitigate 3.2 Gt CO₂‑eq (100-yr)/yr.
Due to the short lifetime of most high-GWP refrigerants, the climate benefit of phasing them out for alternatives is higher on a 20-year time horizon, making this solution highly impactful in the short-term. The use of low-GWP refrigerants currently saves an estimated 0.38 Gt CO₂‑eq (20-yr)/yr. The achievable 20-year impact is 4.7–5.0 Gt CO₂‑eq/yr, with a ceiling of 5.9 Gt CO₂‑eq/yr.
Since medium-GWP refrigerants are less effective at reducing emissions, the climate impacts are lower. If the same achievable adoption scenarios are reached but the effectiveness is calculated for medium-GWP refrigerants in commercial refrigeration, transport refrigeration, and stationary air conditioning applications, the climate impact reduces to 2.2–2.4 Gt CO₂‑eq (100-yr)/yr or 3.9–4.1 Gt CO₂‑eq (20-yr)/yr.
Our findings for impact are higher than many estimates of the scale of current refrigerant emissions. This is because other reports of F-gas emissions typically do not include high-GWP ozone-depleting refrigerants such as CFCs and HCFCs. The IPCC Sixth Assessment (2023) estimates 1.4 ± 0.41 GtCO₂‑eq/yr of 2019 emissions were F-gases from all sources, but this value does not include CFCs or HCFCs. The UNEP (2022) estimates that CFCs and HCFCs stored in equipment produce almost twice as many CO₂‑eq emissions as HFCs do. Our calculated achievable climate impact accounts for all major high-GWP refrigerant chemicals (CFCs, HCFCs, and HFCs), and therefore an achievable climate impact much higher than 1.4 GtCO₂‑eq/yr is reasonable.
Table 7. Climate impact at different levels of adoption of low-GWP alternative refrigerants.
Unit: Gt CO₂‑eq/yr, 100-yr basis
| Current adoption | 0.20 |
| Achievable – low | 2.50 |
| Achievable – high | 2.70 |
| Adoption ceiling | 3.20 |
Transitioning from HFCs to refrigerants with lower GWP can increase jobs (Colbourne et al., 2013; U.S. EPA, 2025). Reports from the Alliance for Responsible Atmospheric Policy and collaborators found that moving toward lower GWP refrigerants in the United States would increase jobs, increase manufacturing outputs of alternative refrigerants, and create more exports, strengthening the United States’ trade position (Inforum et al., 2019; JMS Consulting & Inforum, 2018). It is possible that using alternative refrigerants could lead to consumer savings on energy bills, depending on the alternative refrigerant, application, and equipment design (Colbourne et al., 2013; Purohit & Höglund-Isaksson, 2017; Shah et al., 2019; Zaelke & Borgford-Parnell, 2015). For example, an analysis of mobile air conditioning found that switching to an alternative refrigerant, such as R152a, can lead to high cost savings over its lifetime, and consumers in hotter climates would see the savings benefits (Blumberg et al., 2019). Since efficiency improvements are possible but not guaranteed in all cases, we do not consider this a guaranteed additional benefit.
For a description of the benefits to land resources, please refer to Air Quality below.
Some F-gases such as HFCs are considered per- and polyfluoroalkyl substances (PFAS) and can persist in the environment for centuries, posing serious human and ecosystem health risks (Figure 1) (Dimitrakopoulou et al., 2024; Fenton et al., 2021). PFAS can decompose in the atmosphere to produce trifluoroacetic acid (TFA), which can harm the environment and human health (UNEP, 2023). Possible impacts of high atmospheric TFA concentrations include acid rain, accumulation in terrestrial ecosystems in water and plant matter, and harmful effects on the environment and organisms (Chele et al., 2024; Hanson et al., 2024). Non-fluorinated alternative refrigerants would reduce the amount of PFAS pollution and reduce atmospheric TFA formation, lessening these harmful impacts. Some of these air quality benefits would also benefit indoor air quality because most refrigerants are used in buildings. Using alternative refrigerants avoids release of ozone-depleting substances such as HCFCs that can harm the ozone layer (Bolaji & Huan, 2013).
These benefits depend on the alternative refrigerant used – some low-GWP F-gas refrigerants such as HFOs are highly reactive, can be classified as PFAS, and can form TFA and other degradation products (Salvador et al., 2024). Therefore, the type of alternative refrigerant affects whether this is a benefit or a risk (see Risks below for more information). The thresholds at which these impacts occur are not well understood, and more research is needed to understand the potential harmful effects of TFA (Arp et al., 2024).
Some alternative refrigerants – including propane and ammonia – can react in the atmosphere to form polluting or toxic compounds (Chele et al., 2024). Low- and medium-GWP HFO or HFC refrigerants degrade into TFA, which is considered by some regulating bodies to be a PFAS, a class of chemicals with a proposed ban in Europe (European Chemicals Agency, 2023; European Environmental Bureau, 2025; Garavagno et al., 2024). Although TFA concentrations are currently low and impacts are minimal, increased HFO use could lead to greater accumulation, making it important to further study the potential risks (Chele et al., 2024; European Environmental Bureau, 2025; Hanson et al., 2024; Holland et al., 2021). Moreover, HFOs are made from high-GWP feedstocks, perpetuating the production and release of high-GWP chemicals (Booten et al., 2020; Chele et al., 2024). The use of other alternative refrigerant chemistries will reduce these risks (see Figure 1 and Additional Benefits).
Alternative refrigerants can be flammable (e.g., propane, ammonia) and toxic (e.g., ammonia). This potentially risks the well-being of people or property due to ignition, explosion, or refrigerant leaks (Shah et al., 2017). Minimizing leaks, reducing proximity to ignition sources, enhancing leak sensing, regulating safe charge sizes, and training installation and maintenance professionals are ways to lower this risk (Secop, 2018). Many alternative refrigerants are classified in ASHRAE safety group A2L, and these refrigerants have a low risk of ignition (Gradient, 2015; Imamura et al., 2015). Many countries have updated their standards in recent years to ensure safe use of low-GWP refrigerants, but adoption can be slowed if building codes do not allow for adoption (Heubes et al., 2012; UNEP, 2023).
Some specific technological solutions are required to avoid risks – for example, ammonia corrodes copper (Dräger, n.d.), and CO₂ refrigerant requires equipment and safety mechanisms that can handle its high operating pressure (Zanchi et al., 2019).
Decreasing food loss and waste could require increases in cold storage capacity, especially in commercial, residential, and transport refrigeration (Babiker, 2017; Food and Agriculture Organization of the United Nations, 2019). Alternative refrigerants will lead to reduced GHG emissions from this new food refrigeration equipment, particularly for high-leakage systems such as supermarket refrigeration. However, if less food is produced to better manage food loss, this could lead to a decreased demand for cold storage (Dong et al., 2021).
Alternative refrigerants require design changes (Kim et al., 2020) that could increase the up-front cost of heat pumps.
Using alternative refrigerants will decrease the CO₂‑eq emissions from released refrigerants. This means that management practices to reduce refrigerant release will save fewer CO₂‑eq emissions.
kt high-GWP refrigerant phased out
F-gases
For particular alternative refrigerants and applications, switching to a lower-GWP refrigerant can reduce equipment efficiency (ASHRAE, 2009). Such a switch would decrease direct emissions due to reduction in refrigerant GWP, but would increase emissions associated with electricity generation.
Less efficient refrigerants may also require larger equipment and heavier masses of refrigerants, increasing the emissions for producing and transporting appliances. Fabris et al. (2024) reported that transport refrigeration systems using CO₂ refrigerant are heavier, leading to a 9.3% increase in emissions from fuel consumption during transport.
This map shows the annual average cooling degree days (CDD) for the decade ending in 2025. CDD are a measure of how much the temperature in a location exceeds 21 °C each day, summed cumulatively over a year. Regions with greater cooling degree days will likely have higher demand for space cooling equipment to maintain a comfortable indoor air temperature in buildings and vehicles.
Copernicus Climate Change Service. (2023). ERA5 hourly data on single levels from 1940 to present [Data set]. Copernicus Climate Change Service (C3S) Climate Data Store (CDS). Retrieved January 13, 2026 from Link to source: https://doi.org/10.24381/cds.adbb2d47
Hersbach, H., Bell, B., Berrisford, P., Hirahara, S., Horányi, A., Muñoz‐Sabater, J., Nicolas, J., Peubey, C., Radu, R., Schepers, D., Simmins, A., Soci, C., Abdalla, S., Abellan, X., Balsamo, G., Bechtold, P., Biavati, G., Bidlot, J., Bonavita, M., … Thépaut, J. N. (2020). The ERA5 global reanalysis. Quarterly Journal of the Royal Meteorological Society, 146(730), 1999–2049. Link to source: https://doi.org/10.1002/qj.3803
This map shows the annual average cooling degree days (CDD) for the decade ending in 2025. CDD are a measure of how much the temperature in a location exceeds 21 °C each day, summed cumulatively over a year. Regions with greater cooling degree days will likely have higher demand for space cooling equipment to maintain a comfortable indoor air temperature in buildings and vehicles.
Copernicus Climate Change Service. (2023). ERA5 hourly data on single levels from 1940 to present [Data set]. Copernicus Climate Change Service (C3S) Climate Data Store (CDS). Retrieved January 13, 2026 from Link to source: https://doi.org/10.24381/cds.adbb2d47
Hersbach, H., Bell, B., Berrisford, P., Hirahara, S., Horányi, A., Muñoz‐Sabater, J., Nicolas, J., Peubey, C., Radu, R., Schepers, D., Simmins, A., Soci, C., Abdalla, S., Abellan, X., Balsamo, G., Bechtold, P., Biavati, G., Bidlot, J., Bonavita, M., … Thépaut, J. N. (2020). The ERA5 global reanalysis. Quarterly Journal of the Royal Meteorological Society, 146(730), 1999–2049. Link to source: https://doi.org/10.1002/qj.3803
Deploy Alternative Refrigerants is most effective at mitigating climate change in regions with high levels of current refrigerant use. Unfortunately, there are no comprehensive data available at the country level to estimate the quantity of high-GWP or alternative refrigerants stored in equipment. Countries report HFC stocks and emissions for refrigeration and air conditioning to the UNFCCC, but these do not include high-GWP ozone-depleting substances such as CFCs and HCFCs (UNFCCC, 2023). Since national emissions estimates such as the Emissions Database for Global Atmospheric Research (EDGAR) do not report high-GWP ozone-depleting F-gases, we do not include these data in this analysis (Crippa et al., 2025). The UNEP does track CFC, HCFC, and HFC production and consumption, but this does not provide a comprehensive use or emissions picture since many current emissions are from stock already contained in existing equipment (Ozone Secretariat, n.d.).
Regions with greater cooling demand are likely to require more refrigerant use for refrigeration and air conditioning. Regional patterns of where this solution is most important may evolve in the future as cooling appliances become more widespread and the climate warms.
International, national, and local policies have a large impact on the adoption of alternative refrigerants. The Kigali Amendment to the Montreal Protocol mandates HFC phasedown schedules for participating countries through 2047. Additionally, local building codes and policies influence the use of alternative refrigerants. The Air-Conditioning, Heating, and Refrigeration Institute (AHRI) and The North American Sustainable Refrigeration Council (NASRC) give region-specific information in North America about such policies (AHRI, n.d.-a, n.d.-b; NASRC, n.d.).
CLASP found that in 2025, China and the United States had the highest numbers of both residential air conditioners and refrigerator-freezers (CLASP, 2025). This suggests that residential refrigerant use and emissions are likely to be highest in these countries.
Emissions from producing refrigerants will be higher in locations with more refrigerant manufacturing. Refrigerant manufacturing is more common in locations that are close to chemical feedstocks, have financial incentives, and have experienced and cheap labor (Booten et al., 2020).
Consensus of effectiveness in reducing emissions: High
Phasing out high-GWP refrigerants for low or medium-GWP refrigerants is unquestionably effective at reducing emissions from refrigerant use.
In a report from two U.S. national laboratories, Booten et al. (2020) claim that systems using F-gas refrigerants for refrigeration and air conditioning are “the most difficult and impactful” innovation spaces for refrigerants. Zaelke and Borgford-Parnell (2015) asserted that reducing short-lived climate pollutants including HFCs “is the most effective strategy for constraining warming and associated impacts in the near term.” Utilizing low-GWP alternative refrigerants is a proven means to achieve this.
The IPCC Sixth Assessment (2023) cites the World Meteorological Organization (2018) and Höglund-Isaksson et al. (2017) in claiming that worldwide compliance with the Kigali Amendment schedule would reduce HFC emissions by 61% over baseline emissions by 2050. Velders et al. (2022) modeled future HFC emissions under the Kigali Amendment and found that these HFC reductions could save 3.1–4.4 Gt CO₂‑eq , 100-yr basis/yr by 2050. Dreyfus et al. (2020) estimate possible cumulative savings of 33–47 Gt CO₂‑eq (100-yr) through 2050, with an additional 53 Gt CO₂‑eq (100-yr) through 2060 if HFC phase-down is immediate.
Expert consensus is that the potential impact of alternative refrigerants will increase as a warming climate and increased population and development drive demand for higher use of cooling equipment (Campbell et al., 2018; Dreyfus et al., 2020; Petri & Caldeira, 2015). This will particularly be true for developing countries in already warm climates (Dong et al., 2021).
The results presented in this document summarize findings from one review article, six original studies, two reports, one international treaty, two industry guidelines, one conference proceeding, and eight national GHG inventory submissions to the United Nations. This reflects current evidence from 34 countries, primarily Annex 1 countries as identified by the United Nations as well as China. We recognize this limited geographic scope creates bias, and hope this work inspires research and data sharing on this topic in underrepresented regions.
Deploy Alternative Insulation Materials is defined as using alternative building insulation materials in place of conventional ones. In particular, we highlight the impact of using cellulose instead of glass, mineral, or plastic insulation in new and retrofit buildings. Cellulose insulation manufacture and installation emits fewer GHGs to reach the same operational insulating performance than does manufacture and installation of conventional materials.
Thermal insulation materials are used in the walls, roofs, and floors of buildings to help maintain comfortable indoor temperatures. However, manufacture and installation of insulation materials produces GHG emissions. These are called embodied emissions because they occur before the insulation is used in buildings. Insulation embodied emissions offset a portion of the positive climate impacts from using insulation to reduce heating and cooling demand. A Canadian study found that over 25% of residential embodied emissions from manufacturing building materials can be due to insulation (Magwood et al., 2022). Using cellulose insulation made primarily from recycled paper avoids some embodied emissions associated with conventional insulation.
Insulation is manufactured in many different forms, including continuous blankets or boards, loose fill, and sprayed foam (Types of Insulation, n.d.). Most conventional insulation materials are nonrenewable inorganic materials such as stone wool and fiberglass. These require high temperatures (>1,300 °C) to melt the raw ingredients, consuming thermal energy and releasing CO₂ from fossil fuel combustion or grid power generation (Schiavoni et al., 2016). Other common insulations are plastics, including expanded polystyrene (EPS), extruded polystyrene (XPS), polyurethane (PUR), and polyisocyanurate (PIR). Producing these plastics requires the extraction of fossil fuels – primarily petroleum – for feedstocks, as well as high amounts of energy for processing (Harvey, 2007).
F-gases are often used as blowing agents to manufacture rigid foam board insulation or install sprayed foam insulation. F-gases are GHGs with GWPs that can be hundreds or thousands of times higher than CO₂. High-GWP F-gases used in foam production are released into the atmosphere during all subsequent stages of the foam’s life cycle (Biswas et al., 2016; Waldman et al., 2023). The climate benefits of this solution during the installation stage are primarily due to avoiding these blowing agents.
Alternative insulation is produced from plant or animal biomass (bio-based materials) or waste products (recycled materials). Alternative insulation materials provide climate benefits by consuming less manufacturing energy, using renewable materials in place of fossil fuels, and eliminating high-GWP blowing agents (Sustainable Traditional Buildings Alliance, 2024).
Figure 1 compares a variety of conventional and alternative insulation materials. While many bio-based and recycled materials could be used as alternatives to these conventional materials, this solution focuses on cellulose due to its effectiveness in avoiding emissions, low cost, and wide availability. Cellulose insulation is made primarily from recycled paper fibers, newsprint, and cardboard. These products are made into fibers and blended with fire retardants to produce loose or batt cellulose insulation (Waldman et al., 2023; Wilson, 2021).
Figure 1. Properties and adoption of conventional and alternative insulation materials. Costs and emissions will vary from the values here depending on the insulation form (board, blanket, loose-fill, etc.).
| Category | Material | High-GWP F-gases used? | Median manufacturing and installation emissions* | Mean product and installation cost** | Estimated market share (% by mass) |
|---|---|---|---|---|---|
| Conventional materials | Stone wool | No | 0.31 | 623 | 20 |
| Glass wool (fiberglass) | No | 0.29 | 508 | 34 | |
| EPS | No | 0.38 | 678 | 22 | |
| XPS | Yes, sometimes | 9.44 | 702 | 7 | |
| PUR/PIR | Yes, sometimes | 6.14 | 1,000 | 11 | |
| Alternative materials | Cellulose | No | 0.05 | 441 | 2–13 |
| Cork | No | 0.30 | 1,520 | Commercially available, not widely used | |
| Wood fiber | No | 0.13 | 814 | Commercially available, not widely used | |
| Plant fibers (kenaf, hemp, jute) | No | 0.18 | 467 | Commercially available, not widely used | |
| Sheep’s wool | No | 0.14 | 800 | Commercially available, not widely used | |
| Recycled PET plastic | No | 0.12 | 2,950 | Commercially available, not widely used |
*t CO₂‑eq (100-yr) to insulate 100m² to 1m²·K/W
**2023 US$ to insulate 100m² to 1m²·K/W. We use mean values for cost analysis to better capture the limited data and wide range of reported costs.
Although we are estimating the impact of using cellulose insulation in all buildings, the unique circumstances of each building are important when choosing the most appropriate insulation material. In this solution, we do not distinguish between residential and commercial buildings, retrofit or new construction, different building codes, or different climates, but these would be important areas of future study.
In this solution, the effectiveness, cost, and adoption are calculated over a specified area (100 m²) and thermal resistance (1 m²·K/W). The chosen adoption unit ensures that all data are for materials with the same insulating performance. Due to limited material information, we assumed that insulation mass scales linearly with thermal resistance.
To better understand the adoption unit, a one-story residential building of 130 m² floor area would require approximately 370 m² of insulation area (RSMeans, & The Gordian Group, 2023). For a cold climate like Helsinki, Finland, code requires insulation thermal resistance of 11 m²·K/W (The World Bank Group, n.d.). For a warm climate like Jerusalem, Israel, envelope thermal resistance requirements average 1.1 m²·K/W (The World Bank Group, n.d.). Therefore, depending on the location, anywhere from approximately 4–40 adoption units insulating 100 m² to 1 m²·K/W may be needed to insulate a small single-story home to the appropriate area and insulation level.
Would you like to help deploy alternative insulation? Below are some ways you can make a difference, depending on the roles you play in your professional or personal life.
These actions are meant to be starting points for involvement and may or may not be the most important, impactful, or doable actions you can take. We encourage you to explore, get creative, and take a step that is right for you!
Adams, M., Burrows, V., & Richardson, S. (2019). Bringing embodied carbon upfront: Coordinated action for the building and construction sector to tackle embodied carbon. World Green Building Council, Advancing Net Zero, Ramboll, & C40 Cities. Link to source: https://worldgbc.s3.eu-west-2.amazonaws.com/wp-content/uploads/2022/09/22123951/WorldGBC_Bringing_Embodied_Carbon_Upfront.pdf
Amendment to the Montreal Protocol on substances that deplete the ozone layer. (2016, October 15). Link to source: https://treaties.un.org/doc/Treaties/2016/10/20161015%2003-23%20PM/Ch_XXVII-2.f-English%20and%20French.pdf
Andersen, B., & Rasmussen, T. V. (2025). Biobased building materials: Moisture characteristics and fungal susceptibility. Building and Environment, 112720. Link to source: https://doi.org/10.1016/j.buildenv.2025.112720
Asdrubali, F., D’Alessandro, F., & Schiavoni, S. (2015). A review of unconventional sustainable building insulation materials. Sustainable Materials and Technologies, 4, 1–17. Link to source: https://doi.org/10.1016/j.susmat.2015.05.002
Biswas, K., Shrestha, S. S., Bhandari, M. S., & Desjarlais, A. O. (2016). Insulation materials for commercial buildings in North America: An assessment of lifetime energy and environmental impacts. Energy and Buildings, 112, 256–269. Link to source: https://doi.org/10.1016/j.enbuild.2015.12.013
Cabeza, L. F., Boquera, L., Chàfer, M., & Vérez, D. (2021). Embodied energy and embodied carbon of structural building materials: Worldwide progress and barriers through literature map analysis. Energy and Buildings, 231, 110612. Link to source: https://doi.org/10.1016/j.enbuild.2020.110612
Carbon Removals Expert Group Technical Assistance. (2023, December). Review of certification methodologies for long-term biogenic carbon storage in buildings. European Commission. Link to source: https://climate.ec.europa.eu/system/files/2023-12/policy_carbon_expert_biogenic_carbon_storage_in_buildings_en.pdf
Deer et al. (2007). Alaska Residential Building Manual. Alaska Housing Finance Corporation. Link to source: https://www.ahfc.us/application/files/2813/5716/1325/building_manual.pdf
Esau et al. (2021). Reducing Embodied Carbon in Buildings: Low-Cost, High-Value Opportunities. RMI. Link to source: http://www.rmi.org/insight/reducing-embodied-carbon-in-buildings
The Freedonia Group. (2024). Global insulation report. Link to source: https://www.freedoniagroup.com/industry-study/global-insulation
Fabbri, M., Rapf, O., Kockat, J., Fernández Álvarez, X., Jankovic, I., & Sibileau, H. (2022). Putting a stop to energy waste: How building insulation can reduce fossil fuel imports and boost EU energy security. Buildings Performance Institute Europe. Link to source: https://www.bpie.eu/wp-content/uploads/2022/05/Putting-a-stop-to-energy-waste_Final.pdf
Forestry production and trade. (2023). [Dataset]. FAOSTAT. Link to source: https://www.fao.org/faostat/en/#data/FO
Füchsl, S., Rheude, F., & Röder, H. (2022). Life cycle assessment (LCA) of thermal insulation materials: A critical review. Cleaner Materials, 5, 100119. Link to source: https://doi.org/10.1016/j.clema.2022.100119
Gelowitz, M. D. C., & McArthur, J. J. (2017). Comparison of type III environmental product declarations for construction products: Material sourcing and harmonization evaluation. Journal of Cleaner Production, 157, 125–133. Link to source: https://doi.org/10.1016/j.jclepro.2017.04.133
Global Alliance for Buildings and Construction, International Energy Agency, and the United Nations Environment Programme. (2020). GlobalABC roadmap for buildings and construction: Towards a zero-emission, efficient and resilient buildings and construction sector. International Energy Agency. Link to source: https://www.iea.org/reports/globalabc-roadmap-for-buildings-and-construction-2020-2050
Grazieschi, G., Asdrubali, F., & Thomas, G. (2021). Embodied energy and carbon of building insulating materials: A critical review. Cleaner Environmental Systems, 2, 100032. Link to source: https://doi.org/10.1016/j.cesys.2021.100032
Harvey, L. D. D. (2007). Net climatic impact of solid foam insulation produced with halocarbon and non-halocarbon blowing agents. Building and Environment, 42(8), 2860–2879. Link to source: https://doi.org/10.1016/j.buildenv.2006.10.028
Insulation choices revealed in new study. (2019, June 19). Home Innovation Research Labs. Link to source: https://www.homeinnovation.com/trends_and_reports/trends/insulation_choices_revealed_in_new_study
International Energy Agency. (2023). Building envelopes. Link to source: https://www.iea.org/energy-system/buildings/building-envelopes
International Energy Agency, International Renewable Energy Agency, & United Nations Climate Change High-Level Champions. (2023). Breakthrough agenda report 2023. Link to source: https://www.iea.org/reports/breakthrough-agenda-report-2023
Jelle, B. P. (2011). Traditional, state-of-the-art and future thermal building insulation materials and solutions – Properties, requirements and possibilities. Energy and Buildings, 43(10), 2549–2563. Link to source: https://doi.org/10.1016/j.enbuild.2011.05.015
Kumar, D., Alam, M., Zou, P. X. W., Sanjayan, J. G., & Memon, R. A. (2020). Comparative analysis of building insulation material properties and performance. Renewable and Sustainable Energy Reviews, 131, 110038. Link to source: https://doi.org/10.1016/j.rser.2020.110038
Magwood et al. (2022). Emissions of materials benchmark assessment for residential construction report. Passive Buildings Canada and Builders for Climate Action.
Malhotra, A., & Schmidt, T. S. (2020). Accelerating Low-Carbon Innovation. Joule, 4(11), 2259–2267. Link to source: https://doi.org/10.1016/j.joule.2020.09.004
Mályusz, L., & Pém, A. (2013). Prediction of the learning curve in roof insulation. Automation in Construction, 36, 191–195. Link to source: https://doi.org/10.1016/j.autcon.2013.04.004
Maskell, D., Da Silva, C., Mower, K., Rana, C., Dengel, A., Ball, R., Ansell, M., Walker, P., & Shea, A. (2015, June 22). Properties of bio-based insulation materials and their potential impact on indoor air quality. First International Conference on Bio-based Building Materials, Clermont-Ferrand, France.
McGrath et al. (2023). Embodied carbon and material health in insulation. Healthy Building Network, Perkins&Will. Link to source: https://habitablefuture.org/wp-content/uploads/2024/03/96-Carbon-Health-Insulation.pdf
Naldzhiev, D., Mumovic, D., & Strlic, M. (2020). Polyurethane insulation and household products: A systematic review of their impact on indoor environmental quality. Building and Environment, 169, 106559. Link to source: https://doi.org/10.1016/j.buildenv.2019.106559
Northeast Bio-based Materials Collective 2023 summit proceedings. (2023). Link to source: https://massdesigngroup.org/sites/default/files/file/2024/Northeast%20Bio-Based%20Materials%20Collective%202023%20Summit%20Proceedings.pdf
Petcu et al. (2023). Research on thermal insulation performance and impact on indoor air quality of cellulose-based thermal insulation materials. Materials, 16(15), Article 15. Link to source: https://doi.org/10.3390/ma16155458
Rabbat, C., Awad, S., Villot, A., Rollet, D., & Andrès, Y. (2022). Sustainability of biomass-based insulation materials in buildings: Current status in France, end-of-life projections and energy recovery potentials. Renewable and Sustainable Energy Reviews, 156, 111962. Link to source: https://doi.org/10.1016/j.rser.2021.111962
Riverse. (2024, August). Methodology: Biobased construction materials. Link to source: https://www.riverse.io/methodologies/biobased-construction-materials
RSMeans, & The Gordian Group. (2023, September). Installed cost of residential siding comparative study – September 2023 [Report]. The Brick Industry Association. Link to source: https://www.gobrick.com/content/userfiles/files/RSMeans%20Residential%20Siding%20Comparative%20Cost%20Wall%20System%20Study%20Final%202023-09-15.pdf
SaravanaPrabhu et al. (2021). Comparative analysis of learning curve models on construction productivity of diaphragm wall and pile. IOP Conference Series: Materials Science and Engineering, 1197(1), 012004. Link to source: https://doi.org/10.1088/1757-899X/1197/1/012004
Schiavoni, S., D׳Alessandro, F., Bianchi, F., & Asdrubali, F. (2016). Insulation materials for the building sector: A review and comparative analysis. Renewable and Sustainable Energy Reviews, 62, 988–1011. Link to source: https://doi.org/10.1016/j.rser.2016.05.045
Schulte, M., Lewandowski, I., Pude, R., & Wagner, M. (2021). Comparative life cycle assessment of bio-based insulation materials: Environmental and economic performances. GCB Bioenergy, 13(6), 979–998. Link to source: https://doi.org/10.1111/gcbb.12825
Searchinger, T., Peng, L., Zionts, J., & Waite, R. (2024). The global land squeeze: Managing the growing competition for land. World Resources Institute. Link to source: https://www.wri.org/research/global-land-squeeze-managing-growing-competition-land
Stamm et al. (2022). Chemical and environmental justice impacts in the life cycle of building insulation. Energy Efficiency for All, Healthy Building Network. Link to source: https://informed.habitablefuture.org/resources/research/20-chemical-and-environmental-justice-impacts-in-the-life-cycle-of-building-insulation-report-brief
Sustainable Traditional Buildings Alliance. (2024, March). The use of natural insulation materials in retrofit. Link to source: https://stbauk.org/wp-content/uploads/2024/03/The-use-of-natural-insulation-materials-in-retrofit.pdf
The World Bank Group. (n.d.). Mapping energy efficiency: A global dataset on building code effectiveness and compliance: Country profiles. Link to source: https://www.worldbank.org/content/dam/sites/buildinggreen/doc/building_green_country_profile.pdf
Types of insulation. (n.d.). U.S. Department of Energy. Link to source: https://www.energy.gov/energysaver/types-insulation
Waldman et al. (2023). 2023 Carbon Leadership Forum North American material baselines. Carbon Leadership Forum, University of Washington. Link to source: https://carbonleadershipforum.org/clf-material-baselines-2023/
Wang et al. (2023). Can paper waste be utilised as an insulation material in response to the current crisis. Sustainability, 15(22), Article 22. Link to source: https://doi.org/10.3390/su152215939
Wi, S., Kang, Y., Yang, S., Kim, Y. U., & Kim, S. (2021). Hazard evaluation of indoor environment based on long-term pollutant emission characteristics of building insulation materials: An empirical study. Environmental Pollution, 285, 117223. Link to source: https://doi.org/10.1016/j.envpol.2021.117223
Wilson. (2021). The BuildingGreen guide to thermal insulation: What you need to know about performance, health, and environmental considerations. BuildingGreen, Inc.
Zabalza Bribián, I., Valero Capilla, A., & Aranda Usón, A. (2011). Life cycle assessment of building materials: Comparative analysis of energy and environmental impacts and evaluation of the eco-efficiency improvement potential. Building and Environment, 46(5), 1133–1140. Link to source: https://doi.org/10.1016/j.buildenv.2010.12.002
Sarah Gleeson, Ph.D.
Ruthie Burrows, Ph.D.
James Gerber, Ph.D.
Yusuf Jameel, Ph.D.
Daniel Jasper
Alex Sweeney
Aiyana Bodi
Hannah Henkin
Ted Otte
Amanda D. Smith, Ph.D.
Christina Swanson, Ph.D.
To insulate 100 m² to a thermal resistance of 1 m²·K/W using entirely cellulose insulation in place of the current baseline mix of insulation materials is expected to avoid 1.59 t CO₂‑eq on a 100-yr basis (Table 1). Since many of the avoided emissions are F-gases, the 20-yr effectiveness is higher, avoiding 4.07 t CO₂‑eq per unit of insulation. Effectiveness for this solution measures the one-time reduced emissions from manufacturing and installing insulation. Insulation also reduces the energy used while a building is operating, but those emissions are addressed separately in the Improve Building Envelopes solution.
Conventional insulation effectiveness was considered to be a weighted average effectiveness of the current baseline insulation mix, including a small amount of cellulose insulation currently in use.
The largest contributor to conventional insulation embodied emissions is using high-GWP blowing agents to manufacture or install XPS, PUR, or PIR foam. We assumed the use of F-gas blowing agents for all foams, although these are already being regulated out of use globally (Amendment to the Montreal Protocol on Substances That Deplete the Ozone Layer, 2016) and an unknown amount of low-GWP blowing agents are currently used (such as hydrocarbons or CO₂). Therefore, we anticipate the effectiveness of this solution will decrease as F-gases are used less in the future. We assumed that 100% of blowing agents are emitted over the product lifetime.
Cellulose has the greatest avoided emissions of all of the alternative materials we evaluated (Figure 1). The next most effective materials were recycled PET, wood fibers, and sheep’s wool. Conventional materials like XPS, PUR, and PIR that are foamed with F-gases had the highest GHG emissions. For bio-based materials, we did not consider biogenic carbon as a source of carbon sequestration due to quantification and permanence concerns.
Table 1. Effectiveness at reducing emissions.
Unit: t CO₂‑eq /insulation required to insulate 100 m² to a thermal resistance of 1 m²·K/W, 100-yr basis
| 25th percentile | 0.98 |
| Mean | 1.34 |
| Median (50th percentile) | 1.59 |
| 75th percentile | 1.81 |
Available cost data are variable for all materials, particularly those in early-stage commercialization. The mean cost of purchasing and installing cellulose insulation is less than that of any other conventional or alternative insulation material (Figure 1). Compared with the average cost of conventional insulation, the mean cost savings for cellulose insulation is US$193/100 m² insulated to a thermal resistance of 1 m²·K/W. Since most buildings are insulated over greater areas to higher thermal resistances, these savings would quickly add up. When considering the mean cost per median climate impact, cellulose insulation saves US$121/t CO₂‑eq (100-yr basis), making it an economically and environmentally beneficial alternative (Table 2).
We considered conventional insulation cost to be a weighted average cost of the current baseline insulation mix, including a small amount of cellulose insulation currently in use.
For conventional insulation, material costs of purchasing the insulation are higher than costs for installation (US$540 and US$97, respectively, to insulate 100 m2 to a thermal resistance of 1 m²·K/W). Cellulose has a lower product cost and comparable installation costs to conventional materials. We considered all costs to be up front and not spread over the lifetime of the material or building. For each material type, cost will vary based on the form of the insulation (board, loose, etc.), and this should be accounted for when comparing insulation options for a particular building.
We determined net costs of insulation materials by adding the mean cost to purchase the product and the best estimation of installation costs based on available information. Installation costs were challenging to find data on and therefore represent broad assumptions of installation type and labor. Cost savings were determined by subtracting the weighted average net cost of conventional materials to the net cost of an alternative material. Although we used median values for other sections of this assessment, the spread of data was large for product cost estimates and the mean value was more appropriate in the expert judgment of our reviewers.
Table 2. Cost per unit of climate impact.
Unit: 2023 US$/t CO₂‑eq, 100-yr basis
| Estimate | -121 |
Little information is available about the learning rate for new insulation materials. Mályusz and Pém (2013) evaluated how labor time decreased with repetitive cycles for installing roof insulation. They found a learning rate of ~90%, but only for this specific insulation scenario, location, and material. Additionally, this study does not include any product or manufacturing costs that may decrease with scale.
In general, labor time for construction projects decreases with repetitive installation, including improved equipment and techniques and increased construction crew familiarity with the process (SaravanaPrabhu & Vidjeapriya, 2021). However, Malhotra and Schmidt (2020) classify building envelope retrofits as technologies that are highly customized based on user requirements, regulations, physical conditions, and building designs, likely leading to learning rates that are slow globally but where local expertise could reduce installation costs.
Speed of action refers to how quickly a climate solution physically affects the atmosphere after it is deployed. This is different from speed of deployment, which is the pace at which solutions are adopted.
At Project Drawdown, we define the speed of action for each climate solution as emergency brake, gradual, or delayed.
Deploy Alternative Insulation Materials is an EMERGENCY BRAKE climate solution. It has the potential to deliver a more rapid impact than gradual and delayed solutions. Because emergency brake solutions can deliver their climate benefits quickly, they can help accelerate our efforts to address dangerous levels of climate change. For this reason, they are a high priority.
Manufacturing and installation emissions reductions due to the use of alternative building thermal insulation materials are both permanent and additional.
Permanence
There is a low risk of the emissions reductions for this solution being reversed. By using cellulose insulation instead of inorganic or plastic-based insulation, a portion of the manufacturing and installation emissions are never generated in the first place, making this a permanent reduction. Emissions from high-temperature manufacturing, petroleum extraction, and blowing agent use are all reduced through this approach.
Additionality
The GHG emissions reductions from alternative insulation materials are additional because we calculated them relative to a baseline insulation case. This includes a small amount of cellulose materials included in baseline building insulation. Therefore, avoided emissions represent an improvement of the current emissions baseline that would have occurred in the absence of this solution.
Adoption data are extremely limited for alternative insulation materials. All adoption data and estimates are assumed to apply to both residential and commercial buildings, although in reality the uptake of alternative insulation materials will vary by building type due to differences in structures, climate, use type, and regulations. We assume that future uptake of alternative insulation is used only during retrofit or new construction, or when existing insulation is at the end of its functional lifetime.
European sources report that 2–13% of the insulation market is alternative materials. Depending on the source, this could include renewable materials, bio-based insulation, or recycled materials. In 2018 in the United States, 5% of total insulation area in new single-family homes was insulated with cellulose (Insulation Choices Revealed in New Study, 2019).
To convert estimated cellulose adoption percentage into annual insulation use, we estimated 26 Mt of all installed global insulation materials in 2023 based on a report from The Freedonia Group (2024). We calculated an annual use of approximately 1.7 billion insulation units of 100 m² at a thermal resistance of 1 m²·K/W. Therefore, the median cellulose adoption is 14 million units/yr at 100 m² at 1 m²·K/W, calculated from the median of the 2–13% adoption range.
Since this calculation is based on more alternative materials than just cellulose and is heavily reliant on European data where we assume adoption is higher, this estimate of current adoption (Table 3) is most likely an overestimate.
The little adoption data that were considered in this section are mostly for Europe, and some for the United States. We recognize this limited geographic scope creates bias, and hope this work inspires research and data sharing on this topic in underrepresented regions.
Table 3. Current (2017–2022) adoption level.
Unit: units of insulation/yr installed to insulate 100 m² to a thermal resistance of 1 m²·K/W
| 25th percentile | 9000000 |
| Mean | 13000000 |
| Median (50th percentile) | 14000000 |
| 75th percentile | 17000000 |
Very few data are available that quantify adoption trends. In a regional study of several bio-based insulation materials, Rabbat et al. (2022) estimated French market annual growth rates of 4–10%, with cellulose estimated at 10%. Petcu et al. (2023) estimated the European adoption of recycled plastic and textile insulation, biomass fiber insulation, and waste-based insulation to have increased from 6% to 10% between 2012 and 2020.
When accounting for the calculated current adoption, these growth rates mean a median estimated annual increase of 500,000 insulation units/yr required to insulate 100 m² to a thermal resistance of 1 m²·K/W. The increasing adoption of bio-based insulation decreases the use of conventional insulation materials in those regions.
This adoption trend (Table 4) is likely an overestimate, as it is biased by high European market numbers and based on the likely high estimate we made for current adoption.
Table 4. 2012–2020 adoption trend.
Unit: annual change in units of insulation/yr installed to insulate 100 m² to a thermal resistance of 1 m²·K/W
| 25th percentile | 500000 |
| Mean | 800000 |
| Median (50th percentile) | 500000 |
| 75th percentile | 1300000 |
No estimates have been found for the adoption ceiling of this solution, although we expect it to be high given low rates of current adoption and projected increases in building construction in the coming decades (International Energy Agency [IEA], International Renewable Energy Agency, & United Nations Climate Change High-Level Champions, 2023). Two physical factors that could influence adoption are availability of alternative materials and thickness of insulation.
For cellulose insulation, availability does not seem to limit adoption. The Food and Agriculture Organization of the United Nations (2023) reports that there is a much higher annual production of cellulose-based materials (>300 Mt annually of cartonboard, newsprint, and recycled paper) than the overall demand for insulation globally (>25 Mt annual demand; Global Insulation Report, 2024). However, other uses for cellulose products may create competition for this supply.
Increased thickness of insulation could also be a limiting factor because this would reduce adoption by decreasing building square footage, in particular making retrofits more challenging and expensive. Deer et al. (2007) reported that the average cellulose thermal resistance is similar to mineral and glass wool, and lower than plastic insulations made of polystyrene and other foams. If we assume that 50% of plastic insulation cannot be replaced with cellulose due to thickness limitations, this would represent ~20% of current insulation that could not be replaced without structural changes to the building. Therefore, we calculated the adoption ceiling to be 80% of the current insulation that would be reasonably replaceable, or 140 million units/yr required to insulate 100 m² to a thermal resistance of 1 m²·K/W (Table 5).
Uptake of cellulose insulation could also be limited by its susceptibility to absorbing moisture, limiting its use in wet climates or structures that retain moisture, such as flat roofs. Commercialization of alternative insulation materials beyond cellulose and in many different forms (e.g., board, loose-fill) will increase the adoption ceiling across more building types.
Table 5. Adoption ceiling.
Unit: units of insulation/yr installed to insulate 100 m² to a thermal resistance of 1 m²·K/W
| 25th percentile | N/A |
| Mean | N/A |
| Estimate | 140000000 |
| 75th percentile | N/A |
We found no estimates for feasible global adoption of this solution. Rabbat et al. (2022) estimated the adoption levels of several bio-based insulation materials in France in 2050. For cellulose wadding, this was estimated to be 2.1 times the commercialized volume in France in 2020. Although we do not expect France to be representative of the rest of the world, if the predicted adoption trend holds across the world then we expect low adoption in 2050 to be 2.1 times greater than 2023 adoption. This is 29 million units/yr to insulate 100 m² to a thermal resistance of 1 m²·K/W (Table 6).
The IEA (2023) claims that building envelopes need to have their retrofit rate increase by 2.5 times over the current rate in order to meet net zero targets (2023). This is a reasonable high-adoption scenario. Assuming that more retrofits of buildings occur and greater amounts of alternative insulation are installed in new buildings, we estimate that high future adoption of new insulation could occur at 2.5 times the rate of the low-adoption scenario. This is 73 million units/yr to insulate 100 m² to a thermal resistance of 1 m²·K/W (Table 6).
Adoption will be facilitated or limited by local regulations around the world. Building codes will determine the location and extent of use of cellulose or other bio-based insulation. We expect uptake to be different between residential and commercial buildings, but due to insufficient data, we have grouped them in our adoption estimates.
Table 6. Range of achievable adoption levels.
Unit: units of insulation/yr installed to insulate 100 m² to a thermal resistance of 1 m²·K/W
| Current adoption | 14000000 |
| Achievable – low | 29000000 |
| Achievable – high | 73000000 |
| Adoption ceiling | 140000000 |
The climate impacts for this solution are modest compared to current global GHG emissions. Not all conventional insulations have a high environmental impact due to the use of a wide range of materials, forms, and installation methods as well as the recent adoption of lower-GWP blowing agents. Therefore, the potential for further emissions savings is limited.
We quantified the effectiveness and adoption of cellulose insulation, which has the lowest emissions and, therefore, the highest climate impacts of the insulation materials we evaluated. With high adoption, 1.2 Gt CO₂‑eq on a 100-yr basis could be avoided over the next decade (Table 7).
While we only considered the adoption of cellulose insulation in this analysis, a realistic future for lowering the climate impact of insulation may include other bio-based materials, too. Utilizing a greater range of materials should increase adoption and climate impact due to more available forms, sources, and thermal resistance values of bio-based insulation.
Producing and deploying cellulose and other bio-based insulation requires the use of biomass as a feedstock. Multiple climate solutions, in addition to alternative insulation materials, require biomass, and projected demand across solutions greatly exceeds supply. The deforestation that would be required to meet demand would produce emissions far greater than any mitigation gains from full deployment of these solutions (Searchinger, 2024). In addition to deforestation, there would also be costs and emissions incurred to transport biomass from where it is produced to where it can be processed and used. Thus, the achievable climate impacts presented here is only possible if feedstocks are prioritized for this solution. If feedstocks are instead prioritized for other climate solutions (see Interactions for examples), adoption and impact will be lower for this solution. It is not possible to set all biomass-dependent solutions to high adoption levels, add up their impacts, and determine an accurate combined emissions impact.
Note that we calculated the current climate impact using a current materials baseline that includes a small fraction of cellulose. This means that the reported current adoption impact is a slight underestimate compared with the impacts for replacing entirely conventional insulation with the current amount of cellulose insulation in use.
Table 7. Climate impact at different levels of adoption.
Unit: Gt CO₂‑eq/yr, 100-yr basis
| Current adoption | 0.022 |
| Achievable – high | 0.046 |
| Achievable – low | 0.12 |
| Achievable ceiling | 0.22 |
Some alternative insulations can be cheaper than conventional materials. Although there is large variation in evaluation methods and reported costs, our analysis found that cellulose and plant fibers are cheaper than conventional insulation materials such as stone wool, glass wool, and EPS (Figure 1). Depending on the applicable climate conditions and insulation form, switching to alternative insulation materials can result in cost savings for consumers, including homeowners and business owners.
Conventional insulation materials may contribute to poor indoor air quality, especially during installation, and contribute to eye, skin, and lung irritation (Naldzhiev et al., 2020; Stamm et al., 2022; Wi et al., 2021). Additionally, off-gassing of flame retardants and other volatile organic compounds and by-products of conventional insulation can occur shortly after installation (Naldzhiev et al., 2020). Using bio-based alternative insulation products can minimize the health risks during and after installation (McGrath et al., 2023).
Although there is not a scientifically consistent approach to compare the environmental impacts of conventional and alternative insulation materials, a review analysis of 47 studies on insulation concluded that bio-based insulation materials generally have lower impacts as measured through acidification, eutrophication, and photochemical ozone creation potentials than do conventional materials (Füchsl et al., 2022). Other alternative materials such as wood fiber and miscanthus also tend to have a lower environmental footprint (Schulte et al., 2021). The water demand for wood and cellulose is significantly lower than that for EPS (about 2.8 and 20.8 l/kg respectively compared with 192.7 l/kg for EPS) (Zabalza Bribián et al., 2011). While the limited evidence suggests that the alternative material tends to be better environmentally, there is an urgent need to conduct life cycle assessments using a consistent approach to estimate the impact of these materials.
Cellulose insulation is susceptible to water absorption, which can lead to mold growth in wet or humid environments (Andersen & Rasmussen, 2025; Petcu et al., 2023). Reducing this risk either requires an antifungal treatment for the material or limits adoption to particular climates. The thermal performance of cellulose insulation can decrease over time due to water absorption, settling, or temperature changes, but installing it as dense-packed or damp-spray can alleviate this problem (Wang & Wang, 2023; Wilson, 2021).
Bio-based insulation materials tend to be combustible, meaning they contribute more to the spread of a fire than non-combustible stone or glass insulation. Some bio-based materials are classified as having minimal contribution to a fire, such as some cellulose forms, rice husk, and flax (Kumar et al., 2020). These materials are less likely to contribute to a fire than very combustible plastic insulation such as EPS, XPS, and PUR. Fire codes – as well as other building and energy codes – could limit adoption, risking a lack of solution uptake due to regulatory setbacks (Northeast Bio-Based Materials Collective 2023 Summit Proceedings, 2023).
Additives such as fire retardants and anti-fungal agents are added to bio-based insulation along with synthetic binders, which can lead to indoor air pollution from organic compounds, although likely in low concentrations (Maskell et al., 2015; Rabbat et al., 2022).
Upgrading insulation to lower-cost and lower-emitting alternative materials should increase the adoption of other building envelope solutions as they can be installed simultaneously to optimize cost and performance.
Increasing the manufacturing of cellulose insulation, which contains large amounts of recycled paper, could increase the revenues for paper recycling.
This solution uses wood as a feedstock (raw material), including wood, and crop residues. Because the total projected demand for woody biomass for climate solutions exceeds the supply, not all of these solutions will be able to achieve their potential adoption. This solution is in competition with the following solutions for raw material:
Reducing the demand for conventional insulation products and instead making insulation that produces fewer GHGs during manufacturing would slightly reduce the global climate impact of other industrial manufacturing solutions. This is because less energy overall would be used for manufacturing, and therefore other technologies for emissions reductions would be less impactful for insulation production.
insulation units of 100 m² and 1 m²·K/W
CO₂, F-gas
Bio-based insulation materials including cellulose often have lower thermal resistance than some conventional insulation materials. In particular, bio-based materials may require a thicker layer than plastic insulation to reach the same insulating performance (Esau et al., 2021; Rabbat et al., 2022). Usable floor area within a building would need to be sacrificed to accommodate thicker insulation, which would potentially depreciate the structure or impact the aesthetic value (Jelle, 2011). This would be a more significant trade-off for retrofit construction and buildings in densely developed urban areas.
Sourcing bio-based materials has environmental trade-offs that come from cultivating biomass, such as increased land use, fertilizer production, and pesticide application (Schulte et al., 2021). Using waste or recycled materials could minimize these impacts. Binders and flame-retardants may also be required in the final product, leading to more processing and material use (Sustainable Traditional Buildings Alliance, 2024).
The effectiveness of deploying alternative insulation is not inherently dependent on geographic factors since it addresses emissions embodied in the manufacture and deployment of insulation materials. However, due to a lack of related data, we assumed a consistent global breakdown of currently used insulation materials when in reality, the exact mix of insulation currently used in different geographic locations will affect the emissions impact of switching to alternative materials.
Building insulation is used in higher quantities in cold or hot climates, so deploying alternative insulation is more likely to be relevant and adopted in such climates. Other geographic factors also impact adoption: Areas with higher rates of new construction will be better able to design for cellulose or other alternative insulation materials, and drier climates will face a lower risk of mold growth on these materials. Local building codes, including fire codes, can also affect the adoption of alternative materials.
There are no maps for the Deploy Alternative Insulation Materials solution. It is intended to address emissions embodied in the manufacture and deployment of insulation materials and has no intrinsic dependence on geographic factors.
Consensus of effectiveness in reducing building sector emissions: Mixed
There is scientific consensus that using building insulation with lower embodied emissions will reduce GHG emissions, but expert opinions about the magnitude of possible emissions reductions as well as the accuracy of determining these reductions are mixed.
Biswas et al. (2016) determined that, for insulation, avoided emissions from reduced heating and cooling energy tend to outweigh the embodied emissions. However, others emphasize that as buildings become more energy-efficient, material embodied emissions become a larger factor in their carbon footprint (Cabeza et al., 2021; Grazieschi et al., 2021). Embodied emissions from insulation can be substantial: Esau et al. (2021) analyzed a mixed-use multifamily building and found that selecting low-embodied-carbon insulation could reduce building embodied emissions by 16% at no cost premium.
Multiple studies have found that some sustainable insulation materials have lower manufacturing emissions than traditional insulation materials (Asdrubali et al., 2015; Füchsl et al., 2022; Kumar et al., 2020; Schiavoni et al., 2016). However, researchers have highlighted the difficulty in evaluating environmental performance of different insulation materials (Cabeza et al., 2021; Grazieschi et al., 2021). Gelowitz and McArthur (2017) found that construction product Environmental Product Declarations contain many errors and discrepancies due to self-contradictory or missing data. Füschl et al. (2022) conducted a meta-analysis and cautioned that “it does not appear that a definitive ranking [of insulation materials] can be drawn from the literature.” In our analysis, we attempted to compare climate impact between materials, but we acknowledge that this can come from flawed and inconsistent data.
Despite the difficulties in comparing materials, there is high consensus that cellulose is a strong low-emissions insulation option due to its low embodied carbon, high recycled content, and good thermal insulating performance (Wilson, 2021).
The results presented in this document summarize findings from four reviews and meta-analyses, 14 original studies, three reports, 27 Environmental Product Declarations, and two commercial websites reflecting current evidence from eight countries as well as data representing global, North American, or European insulation materials. We recognize this limited geographic scope creates bias, and hope this work inspires research and data sharing on this topic in underrepresented regions.
Join the 85,000+ subscribers discovering how to drive meaningful climate action around the world! Every other week, you'll get expert insights, cutting-edge research, and inspiring stories.