DECARBONISATION
Harnessing the AI boom for NHS Net Zero
Steven Heape explores how the explosive growth in AI and data centre demand could hand the NHS the revenue generator it has lacked since CHP fell out of favour, and sets out what a genuinely investable model for combined heat and compute might look like.
Since 2020, the NHS has benefited from around £1.5bn of Public Sector Decarbonisation Scheme (PSDS) funding, delivered via Salix, to install heat decarbonisation technology across the estate. That funding has been transformational. It has proven that heat pumps, geothermal pilots and large-scale retrofits can work at NHS scale, and it has built a supply chain capable of delivering them. The NHS has a target to reach Net Zero for building emissions by 2040, ten years ahead of the government’s own 2050 target for the country as a whole. But PSDS funding has now been withdrawn, and the fundamental question it leaves behind has not gone away: how does the NHS continue this journey without capital support or revenue funding to soften the cost? The government and NHS strategy both lean heavily
on the fact that the electricity grid is decarbonising at a national, centralised level. That is genuinely good news: it means we do not need to solve electricity decarbonisation site by site. But it also means heat is now the hard part. We need to move away from natural gas towards electricity- based or renewable heat solutions, and heat pumps have been seen as the principal route to do this.
The temperature problem, again For new buildings built to the Net Zero Hospital standard, this is genuinely achievable. Low-temperature heating circuits, designed from the outset around heat pumps producing heat at around 50°C, can achieve coefficients of performance (COP) in the fours or fives, genuinely efficient and genuinely deliverable. For the majority of our acute and community mental
health estate, the picture is much harder. Retrofitting an entire hospital to operate comfortably at 50°C is prohibitively expensive, and in the absence of public capital, that expense has nowhere obvious to go. What we do have is a huge amount of what funders call patient capital: money that institutional investors want to deploy over 25, 30 or even 40-year horizons. But patient capital is not free money. It needs something that pays it back. And the NHS has to think carefully about its balance sheet: it cannot simply take on new capital commitments without a genuine off-balance-sheet mechanism to support them.
Why CHP stopped working Ten to fifteen years ago, combined heat and power (CHP) gas engines were the default answer to exactly this problem. Even a modest 1–2MW CHP engine could generate revenue savings substantial enough that, over a 15-year life, it would cover the cost of a full energy centre
replacement, stripping out end-of-life steam distribution networks, and still leave the Trust with several million pounds of upside. That model no longer works. As the electricity grid has decarbonised, CHP now emits more carbon than it saves across almost the whole country, bar a handful of areas with particularly carbon-intensive local generation. NHS England has, for good reason, moved away from it. Which leaves a real gap: we still have a genuine funding mechanism problem, and we still have to decarbonise the heat. There is a more direct alternative worth confronting
head-on: simply converting a hospital’s heating to a full heat pump system. For a medium-sized acute hospital, that is typically in the order of £12m of capital investment, and because electricity remains several times more expensive than gas per unit of delivered heat even at reasonable coefficients of performance, it also carries a negative impact on the revenue budget of around £500,000 a year. In plain terms: unaffordable, for most Trusts, without external support. The alternatives that exist today do not fully close that gap. Deep geothermal is real and can work, but it is highly locationally specific, and comes with substantial upfront development risk, typically £2–2.5m just to establish whether a suitable resource exists, before a project of perhaps £25m even begins construction. Over a 40-year horizon that can wash its face and suit patient capital well, but the early risk is a genuine barrier for many Trusts. Large-scale solar PV faces a different constraint: land.
There has been grant funding available for PV, but it has been comparatively small, and the national-level intent, that PV should reduce net electricity costs enough to make heat pumps genuinely affordable, runs into a structural problem: until a mechanism exists to ring-fence and recirculate the revenue savings PV actually generates, Trust finance teams understandably take that benefit and divert it to pressing clinical needs, rather than reinvesting it into the heat pump projects it was meant to enable. The scale required also deserves to be stated plainly. A medium-sized acute hospital using around 20GWh
November 2026 Health Estate Journal 31
The NHS aims to reach Net Zero emissions by 2040.
AdobeStock / IzzetNoyan
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72
