IN SUMMARY
Heat pumps are established as a low-carbon alternative to gas boilers, with Gemserv’s research finding a gas boiler generates 55% more emissions in use than a 10kW air source heat pump. But the emissions embodied in making a heat pump are increasingly under scrutiny from regulators, investors and customers. Minviro and Gemserv ran a cradle-to-gate life cycle assessment of a 10kW air source heat pump to measure those embodied emissions, identify the hotspots, and show how far they can be reduced.
Manufacturing carries a real carbon cost. A 10kW air source heat pump made with standard modern methods has 640 kg CO₂e of embodied emissions, before its clean in-use performance is counted.
Better choices nearly halve it. Switching to lower-carbon steel, recycled insulation and a lower-GWP refrigerant cuts embodied emissions by 46.8%, to 340 kg CO₂e.
Four hotspots dominate. Steel production, tube insulation, on-site manufacturing electricity and refrigerant choice account for the bulk of the footprint, and each offers a clear reduction pathway.
Why manufacturing emissions matter for heat pumps
The case for heat pumps rests on their in-use performance, and that case is strong: a gas boiler generates 55% more emissions in operation than a comparable 10kW air source heat pump. But as clean technologies scale, attention is shifting to their whole-life footprint, including the materials and energy used to build them. The consequences of ignoring this are visible across the wider sector, from landfilled first-generation wind turbines to water stress from lithium mining. Regulation is following, with the Ecodesign for Sustainable Products Regulation, the Corporate Sustainability Reporting Directive and the EU and UK Taxonomies all drawing on life cycle thinking. For heat pump makers, the in-use savings are only fully realised if manufacturing is decarbonised too.
How the study was built
Minviro conducted a cradle-to-gate LCA with a functional unit of one 10kW air source heat pump ready for shipping, covering raw material extraction through to manufacturing. Transport, the use phase and end-of-life were excluded so the focus stayed on the materials and processes of production. The bill of materials drew on established academic studies, with background data from Ecoinvent 3.10 and the Environmental Footprint 3.1 method, focused on climate change potential. Importantly, the base model uses global average emission factors for all inputs, making the results location-agnostic. They are an illustrative example for discussing reduction opportunities rather than a figure for any one product, since every supply chain differs and any product-specific claim needs its own regionally representative LCA.