There’s hundreds of innovative energy technologies out there competing for attention and investment. What really matters?
I always find plenty of food for thought in the regular Innovating to Net Zero reports from the UK’s Energy Systems Catapult. The 2026 edition is no exception.
These reports are wide-ranging studies of how innovative energy generation, transmission and storage technologies can come together to help the UK reach its goals of (somewhat hedged) clean power generation by 2030 and net zero greenhouse gas emissions by 2050. They’re also a great example of what the publicly-funded Catapults can do, in terms of providing strategic leadership on limited resources.
The 2024 report highlighted the challenge of those times when demand outstrips supply – like those winter periods when the sun doesn’t shine, the wind doesn’t blow, but everyone wants to put the heating on.
This year’s report focuses on closing those gaps over different timescales. There’s the predictable peaks and troughs over the usual course of a day, and the less predictable weather-dependent swings over longer periods. For example, how much and what kind of energy storage do we need to defend ourselves against the dreaded dunkelflaute?
It’s no surprise that the solution lies in the right mix of renewables, nuclear and flexible grid management. Many other reports from the ES Catapult and other bodies have come to the same conclusion. But here’s a few points from the new report that struck me as interesting.
Renewables AND nuclear
An optimal mix of low-carbon power generation technologies could cut the cost of the energy transition by at least £70 billion, compared to a “steady transition” scenario with low levels of renewables.
As usual, the scenarios with the biggest and quickest deployment of renewables have the lowest system costs. But that doesn’t mean there’s no need for nuclear. Nuclear and renewables are extremely complementary – they work well together in a low-carbon energy system, even if the two industry’s loudest voices often don’t seem to be in accord.
Alongside the continuing roll-out of offshore wind and onshore solar, a rapid scale-up of new nuclear power through the 2030s is an essential part of the mix. The Catapult forecasts a role for over 25GW of nuclear, both large-scale and small modular reactors (SMRs), by 2040. That would run close to baseload for most of the time, significantly reducing some of the most challenging gaps.
Building new nuclear plant on this schedule is the big challenge for the industry, as it has been since the UK mooted a new generation of nuclear two decades ago. The report urges an acceleration of the first SMRs to as early in the 2030s as possible, with the government still needing to set a clear route to market.
With no new nuclear, the system costs could increase by up to £50 billion by 2050. Gas-fired power stations, with or without carbon capture, would have to fill the gap.
Carbon captured
Carbon capture and use (CCS) is still unproven at scale, but the Catapult expects it to be used extensively in industries such as cement which are otherwise hard to decarbonise, as well as “blue” hydrogen production.
Using CCS to allow the continued use of fossil fuels for power generation is likely to be prohibitively expensive to use at scale, however. It may still have a small role in small-scale rapid-response power plant which only fire up when demand is peaking – maybe 4GW in total by 2040.
Unabated gas-fired power stations will still be used in 2040 and beyond, but only to meet peaks of demand. It’s hard to predict how much capacity will be needed – maybe at least 20GW by 2040, but that would be used decreasingly often until it’s well and truly obsolete. If gas-fired power stations are only needed for a few days a year, however, they will need to be financed in a very different way.
Blues for hydrogen
Hydrogen turbines could be a more cost-effective option for peaking plant. The Catapult sees “blue” hydrogen, produced from natural gas by steam methane reforming with CCS, as the key to large-scale production. Capturing the carbon from continuous production should be much more efficient than intermittent capture from a fossil-fired power station.
Blue hydrogen production is still unproven at scale. If it doesn’t meet its cost targets, or CCS just doesn’t work efficiently on the required scale, then there will be more demand for other hydrogen technologies.
The obvious replacement would be “green” hydrogen, produced by electrolysis using low-carbon electricity, which otherwise may be limited to niche industrial roles. Longer-term, high-temperature nuclear reactors could be more efficient.
Battery in your driveway
By putting a battery in every driveway, electric cars can do a lot to help smooth electricity supply and demand. The Catapult reckons it’s enough to provide all the extra flexibility needed over a day.
There’s two ways to exploit car batteries. The first is smart charging, where the cars only charge when power is plentiful and cheap. That’s already used by many EV owners. The more challenging option is vehicle-to-grid, where cars can top up at times of low demand, then feed back into the grid when needed.
There’ll be a few challenges to sell that to car owners. There’s the worry that your battery will be emptied just when you need it. And people will need to keep their cars plugged in whenever they’re parked at home, which would be an unwelcome faff if you’re in and out several times over the day. That’s all if you’re lucky enough to have your own drive or regular parking spot, of course.
Big data, big challenge
Integrating a lot of small-scale smart technologies will demand a lot of digital monitoring, management and control systems. New technologies will be needed to manage the flow of information for real-time grid management, forecasting and coordination across potentially millions of distributed and diverse assets.
There’s a big role for digital technologies which might be called AI. The actually useful ones are, of course, very different to the LLM-type generative platforms which most people will think of.
Last month, the ES Catapult promoted the State of AI for Decarbonisation 2025 report. It was notable that almost all of the actual achievements described therein were based on good old machine learning and other technologies that generally wouldn’t have been called AI before the current hype cycle.
Independent energy analyst Ketan Joshi recently made the same point in a widely-discussed paper on AI tech greenwashing.
The new data centres being built for those generative AI technologies will add significant extra load to the energy system – potentially 28TWh a year by 2040, around a third of total industrial demand, according to the ES Catapult. That’s a huge additional burden for something of rather questionable value.
The map and the territory
A welcome new feature in this report is a System of Systems Map which shows how all these different energy technologies, markets and organisations fit together.
It’s a useful tool for thinking about how they can be integrated to improve overall efficiency and reliability, while lowering costs and avoiding dead ends.
It’s also a handy starting point for innovative companies working in the clean energy space to understand where they fit, identify stakeholders, and spot strategic opportunities.
Check out the full report and other resources here.