Waste heat recovery is the process of capturing excess thermal energy from industrial operations and redirecting it for productive use. Instead of letting heat escape through flue gases or cooling water, these systems channel it into nearby processes, power generation, or local heating networks. It can lower fuel bills and cut emissions, though its success depends heavily on the temperature of the heat source and its distance from a user.
The technology is not a universal fix. Much waste heat is too cool, too far from anyone who could use it, or too costly to capture, so it dissipates into the atmosphere. Understanding those limits explains why some recovery projects go ahead while others stop at the feasibility stage.
How it works
Waste heat, also known as surplus or residual heat, is energy generated during an industrial process that is not used and is instead released into the environment. This heat travels through various flows, such as hot flue gases, exhaust air, or cooling fluids. The UK government’s Industrial Heat Recovery Support (IHRS) programme defined it as heat “generated in or for an existing industrial process but is currently emitted to the environment”.
To capture this energy, engineers use specific technologies depending on the temperature of the waste stream. The US Department of Energy (DOE) groups these sources into three bands:
- High temperature: Above 650 °C (1,200 °F). This includes steel electric arc furnaces and nickel refining furnaces running at 1,370–1,650 °C. The heat is typically used for combustion-air preheating or steam and power generation.
- Medium temperature: Between 230 °C and 650 °C (450–1,200 °F).
- Low temperature: Below 230 °C (450 °F). This includes air compressors running at 30–50 °C, engine cooling at 70–120 °C, and annealing furnace cooling at 70–230 °C.
Recovering low-temperature heat is often more difficult. The DOE notes that capturing heat from low-temperature combustion exhausts is often impractical because acidic condensation corrodes heat exchangers. In other cases, heat pumps, the same technology used in ground-source geothermal heating, can raise low-grade heat to a useful temperature, though, as the DOE puts it, their efficiency decreases as the desired temperature lift increases.
Real-world examples
Several projects show how waste heat recovery systems function in industrial settings, while others show why some projects are abandoned.
Wienerberger Broomfleet brickworks (East Yorkshire) Funded by the IHRS programme, this project went into full operation in August 2022. A heat exchanger on the kiln exhaust delivers pre-heated air to the dryer and to the kiln combustion air. According to the government case study (2023), the system recovers about 525 MWh a month and cut the stack temperature from 230 °C to 140 °C.
Pilkington (NSG Group), Lathom, Lancashire Not all recovery projects go ahead. An IHRS-funded feasibility study (2021) found potential at the Pilkington glass site to cut gas use by 9.6%, worth over £25,000 a year. However, the study concluded that the “only viable technical solution … is not commercially viable”, and the project was suspended. Data centre heat, London (Old Oak and Park Royal) In November 2023 the government’s Green Heat Network Fund awarded £36m to the Old Oak and Park Royal Development Corporation. The planned network is to heat more than 10,000 new homes and 250,000 m² of commercial space with waste heat from nearby data centres, which the government described as a first for the UK. Trade press reports that heat pumps will upgrade the low-grade data-centre heat, which sits at about 20–35 °C. We could not confirm whether construction has started, so treat this as a funded, planned scheme.
Data centre heat, Finland (Microsoft and Fortum) Announced in March 2022, the plan is for Microsoft’s data centres in Espoo and Kirkkonummi eventually to supply approximately 40 percent of the district heat needed in the area, according to Eurelectric, using about 75% of the data centres’ waste heat each year. Fortum’s network serves about 250,000 users. We could not confirm whether the heat is flowing yet.
Japan steelworks (1999) A steelworks in Japan producing 6 million tonnes a year used a Kalina cycle to generate power from cooling water at 98 °C, according to the DOE. This shows that ammonia–water cycles can be applied in heavy industry.
What the numbers say
A large amount of heat is lost, but the commercially viable portion is much smaller.
How much energy is lost? The US DOE (2008) estimates that between 20% and 50% of industrial energy input is lost as waste heat, while noting that the exact quantity is poorly quantified. It puts unrecovered US industrial waste heat at anywhere from 5 to 13 quadrillion Btu per year. In the UK, a 2014 analysis by Element Energy for DECC, cited by the government, found that about 48 TWh/yr of energy is lost as heat in industrial processes across eight energy-intensive sectors.
How much can actually be recovered? Not all of this heat is accessible. Of the 48 TWh/yr lost in the UK, 11 TWh/yr is technically recoverable, meaning a suitable heat user exists within 40 km. Only 7 TWh/yr has economic potential, around 2% of overall UK industrial heat use. Just 5 TWh/yr is commercially viable, defined as having a payback period of under two years. Most waste heat is too cool, too far from a user, or too costly to capture.
Technologies and efficiency
- Recuperators and regenerators: These are heat exchangers that preheat combustion air. Recuperators pass heat continuously through a wall, while regenerators alternately heat and cool a thermal mass, such as brick checkerwork or a rotating wheel, to store heat.
- Economisers and waste heat boilers: Economisers recover flue-gas heat into boiler feedwater; waste heat boilers use it to raise steam.
- Steam Rankine cycle: The DOE calls this the most efficient option for heat above about 340–370 °C (650–700 °F).
- Organic Rankine Cycle (ORC): This uses an organic fluid with a lower boiling point, allowing it to run on waste heat sometimes as low as 66 °C (150 °F). The DOE notes that an ORC efficiency of 10–20% is a substantial percentage of theoretical efficiency at these low temperatures.
- Kalina cycle: This uses an ammonia–water mixture. The DOE states it has proven successful for recovering low- to medium-temperature waste heat.
- Thermoelectric generators (TEGs): These generate electricity directly from a temperature difference. Contrary to claims that they are highly efficient, the US DOE (2008) reports that most systems in use have efficiencies of only 2–5%. They are mainly used on spacecraft and in remote locations, and had not been demonstrated in industrial applications at the time of that report. We cover them in more detail in our article on thermoelectric generators for waste heat recovery.
Costs Indicative installed costs from the DOE (2008) were:
- Kalina cycle: $1,100–1,500 per kW electric.
- ORC: $1,500–3,500 per kW electric.
These figures are from 2008 and should not be treated as current prices.
Market size The only market figures available come from commercial research firms, and estimates vary widely because firms define the market differently. ReportCubes puts the US market at US$13.6bn in 2025, Global Market Insights puts North America at US$18.7bn in 2025, and Meticulous Research puts the global market at US$67.42bn in 2025. We found no checked figure for a separate “electricity/steam generation” segment.
Limitations / honest assessment
Waste heat recovery is not a silver bullet. The DOE lists several barriers to wider adoption:
- Distance and temperature: Heat is difficult to transport, and much of it has no nearby end use. This is why only a fraction of technically recoverable heat is ever recovered.
- Corrosion and maintenance: Recovering heat from low-temperature flue gases can cause acidic condensation. Material constraints and costs, along with maintenance costs, weigh on the economics.
- Heat-to-power limits: Converting waste heat into electricity is inefficient, especially at low temperatures. As a general rule, using the recovered heat directly is usually the better option.
- Payback periods: Payback is not reliably short. In the UK figures, only 5 of 11 TWh/yr of technically recoverable heat was commercially viable (payback under two years), and the DOE lists long payback periods as a barrier.
The UK angle
The UK has supported heat recovery through government funding schemes, though both main schemes have now closed.
Industrial Heat Recovery Support (IHRS) This programme in England and Wales had an £18 million pot, allocated competitively. It was split into £6m for feasibility studies (Phase 1) and £12m for capital grants (Phase 2), with capital grants capped at £1.5m per project. It closed to applications in July 2020 and completed in March 2022. The government lists 23 named case-study organisations, including British Sugar, Cargill and Wienerberger.
Industrial Energy Transformation Fund (IETF) The IETF, available in England, Wales and Northern Ireland, made “up to £500m” available with funding until 2028. According to the government’s benefits report (2026), as of October 2025 it had supported over 150 projects. A total of £201.5m in grants was offered, well below the £500m ceiling, leveraging £430.7m of private funding. Eligibility was strict: heat recovery was supported only when the recovered energy was used in the site’s own industrial processes, not for exporting heat to other sites or for space heating. The government has said there will be no further extension of the IETF, and no successor fund is planned.
Heat networks The government estimates that heat networks supply about 3% of UK heat but could supply about 20% by 2050. It has set an interim target for England of 7% (27 TWh) of heat demand by 2035. The Energy Act 2023 gave powers for heat network zoning in England; we could not confirm whether the zoning regulations are in force. Power generated from waste heat displaces grid supply, whose live mix and carbon intensity we track.
Frequently asked questions
What is waste heat recovery, and how does it work?
Waste heat recovery is the process of capturing excess thermal energy from industrial operations, such as flue gases or cooling water, and redirecting it for productive use. It works by using heat exchangers, heat pumps, or heat-to-power cycles to transfer this energy to another process, power generation, or a local heating network.
What are the main types of waste heat recovery system?
The main systems include recuperators and regenerators (which preheat combustion air), economisers and waste heat boilers (which heat boiler feedwater or raise steam), heat pumps, and heat-to-power systems. Key heat-to-power technologies are the steam Rankine cycle, the Organic Rankine Cycle (ORC) for lower temperatures, and the Kalina cycle. Thermoelectric generators convert temperature differences directly into electricity, but the DOE (2008) puts most at only 2–5% efficient.
How much industrial energy is lost as waste heat?
The US DOE estimates that between 20% and 50% of industrial energy input is lost as waste heat. In the UK, about 48 TWh/yr is lost as heat in industrial processes (2014 figures), of which only 7 TWh/yr has economic potential and 5 TWh/yr is commercially viable.
Can waste heat from data centres heat homes in the UK?
It is planned. The Old Oak and Park Royal scheme in London has £36m of government funding to heat more than 10,000 new homes and commercial space with waste heat from data centres, but we could not confirm that construction has started. Microsoft and Fortum are pursuing a similar plan in Finland.
Is there UK government funding for industrial heat recovery?
Not at present. The Industrial Heat Recovery Support (IHRS) programme closed to applications in 2020 and ended in 2022. The Industrial Energy Transformation Fund (IETF) offered £201.5m in grants against a ceiling of up to £500m, has closed to new rounds, and no successor is planned. Check GOV.UK for any new schemes.
How big is the waste heat recovery market?
There is no single reliable figure. Research firms put it at US$13.6bn for the US (ReportCubes) and US$18.7bn for North America (Global Market Insights) in 2025, while Meticulous Research gives US$67.42bn globally. Estimates vary widely because firms define the market differently.