The A to Z of the Energy Transition: N is for Nuclear Energy

Few topics in energy are as emotive and polarised as nuclear. For some it is a major and growing source of large scale, reliable, 24/7 low-carbon power. For others, it is an expensive, outdated and unreliable form of energy.

As a disinterested party (which is not the same as uninterested!), I think it's fair to say there are elements of truth to both of those seemingly conflicted statements. I'll do my best to share some perspectives on the role of nuclear today and its potential role in the future, as well as emerging new technologies. But first a brief bit of history.

History of nuclear energy

Although discovery of radiation goes back to the 19th Century, it was the period around World War Two that recognition of the role that nuclear could play in energy started. The U.S. Atomic Energy Commission was established in 1946 to oversee nuclear energy development and in 1951 the first experimental nuclear reactor (EBR1-1) in Idaho successfully produced electricity. In June 1954, the Obninsk Nuclear Power Plant in the former Soviet Union became the world's first nuclear power plant to generate electricity for the grid (just 5MW - equivalent to a large single onshore wind turbine today). And it was in the UK the first commercial nuclear power station came online in 1965, at Calder Hall, Cumbria, with 65 MW capacity (tiny compared by today's standards).

Nuclear scaled dramatically in the decades that followed, but only in countries with access to nuclear technology - notably the US, former Soviet Union (today's Russia and Ukraine, Canada, France, India, Japan, South Korea and the UK).

However, since 2000 global nuclear production has been broadly flat. As new plants came online, particularly in China and India, capacity in Europe has fallen - notably the UK has retired aging plants and and Germany (for political reasons) shutdown its last plant in 2023. After nearly 60 years of nuclear electricity, Germany produced precisely 0 TWh in 2024, although remained reliant on nuclear from its French neighbour.

2024 saw nuclear generation reach a new all-time high of 2817 TWh, although this is only 9% higher than 2000. The US remains the largest producer of nuclear energy, at over 823 TWh in 2024, versus 451 TWh in China and 380 TWh in France. France also retained the crown for the highest proportion of nuclear, with nearly 70% of its electricity nuclear powered - far higher than any other economy (the US is 18%, UK 14% and globally around 9%).

Attributes of nuclear energy

Nuclear has some important attributes.

First, it provides 24/7 stable electricity, so is often seen as providing 'base load' to a grid. Some question whether base load is still the right way to think about grids, but having a stable supply of electricity is generally a good thing. However, in contrast to gas generation or hydro, nuclear has less flexibility. It cannot quickly be ramped up or ramped down to respond to demand.

Nuclear is often seen as having the lowest full life-cycle CO2 emissions of any energy, typically lower than wind and solar. It is also almost the safest form of energy, only beaten by solar (although I would be surprised if this was still the case if the full solar supply chain was considered), based on Our World in Data's excellent chart below (which also shows emissions). Of course, in common with other low frequency, high impact events (such as airline accidents), when it goes wrong it is a very big story. I don't intend to dwell on the very few but very significant accidents that have occurred in nuclear (a quick search only lists eight since the 1950s), but it would be naïve to not view this as having had a material impact on the growth (or lack thereof) in nuclear energy. After Japan's 2011 Fukushima, Germany's Government decided to end all nuclear production (as mentioned above closing its remaining units in 2023).

Despite the huge size of Giga Watt scale nuclear parts, no other technology (except potentially fusion in the future) offers such a dense form of energy. Hinckley Point C in Somerset, England when operational will produce 3.2 GW of electricity (around 7% of the UK's power needs) from a single site. Although sites are constrained by needing access to large volumes of cooling water, so typically located on the coast and away from centres of population.

So, nuclear is amongst the safest, lowest carbon forms of energy, with 24/7 output. What's not to like and why hasn't it grown more in recent decades?

Well unfortunately the track record of project delivery has not been stellar. I recently spotted the chart below in the FT, which gives some idea of how much nuclear project costs and schedules have over run. For example, Flamanville 3 in France overran by more than a decade and its final cost was over 3x higher than the initial budget (according to publicly available data), leading to the eventual demise of French developer Areva. The Barakah 1-4 projects in the UAE, based on the same South Korean APR-1400 (Advanced Power Reactors 1400 MW) units as the Saeul 1-2 projects (in South Korea), is a better example, but still over ran on both schedule and budget. China is perhaps the one place where recent nuclear projects have been delivered on schedule and budget, although there is less public data to substantiate this.

All of this has led to nuclear being far more expensive, certainly in Europe and the US, than was historically the case. ICIS, a data and analytics company, estimates that even if the UK's Sizewell C's costs are reduced to £40 billion, the electricity generated could cost £170-£186 per MWh over its lifespan. At these cost levels people, fairly, challenge whether renewables + storage or abated CCGT (combined-cycle gas turbines) could deliver the same low-carbon electricity at a lower cost. For now, at least, the UK has decided that strategically new large-scale nuclear is critical for energy security and has chosen to support the build of Sizewell C. We'll have to wait and see how much the yellow blob on the chart moves.

The future - a nuclear renaissance?

Over the last couple of years, nuclear has been getting a lot more attention. At COP28, a pledge was reached by several countries to triple nuclear by 2050. “If you want to reconcile jobs creation, strategic autonomy and sovereignty and sovereignty, and low carbon emissions, there is nothing more sustainable and reliable than nuclear energy,” said President Emmanuel Macron. Nuclear has also been gaining traction with large data companies, particularly in the US. Google and Microsoft have signed PPAs (Power Purchase Agreements) for nuclear energy to meet increasing data centre demand.

There are also significant developments in nuclear technologies. SMRs (Small Modular Reactors) are compact reactors typically producing around 300 MW. Their modular design allows for factory fabrication and on-site assembly, significantly reducing construction time and cost. This makes them ideal for remote locations, industrial applications (like data centres), and integration with renewable energy systems. SMRs are still sizeable projects, but nowhere near the scale of a multi Giga Watt project. Both China and Russia have already developed protype SMR projects. And Rolls-Royce SMR has been selected to build the UK's first SMR projects (Rolls-Royce SMR will build Britain’s next generation of nuclear power plants).

For SMR to scale at pace, it will be critical to demonstrate that there is a learning curve which brings down costs (or at least doesn't make them increase). Initial estimates of ~$60 per MWh in the US just a few years ago have already increased to ~$90 per MWh (which would still be pretty reasonable). Governments and regulators have a critical role here in trying to standardise policy and regulation across regions, so one design can be scaled and deployed globally, rather than tailored to each location.

Where SMRs typically use third generation water-cooling systems, there are advances in fourth generation technologies typically under the category of AMR (Advanced Modular Reactors). A wide range of cooling technologies, including gas, molten salt, or liquid metal—and a variety of fuel types, including uranium oxide, mixed oxide (MOX), and thorium. These technologies enable higher operating temperatures, improved thermal efficiency, and the potential for applications beyond electricity generation, such as industrial heat. There are also advances in safety that rely on passive (or fail safe) safety systems.

So will nuclear triple by 2050?

Not at the current rate. Again, according to the Energy Institute Statistical Review of World Energy) nuclear energy grew by 2.6% globally in 2024. It needs to grow at around 4% each and every year out to 2050 to triple. Even China, which has grown at 13% pa over the last decade, only managed 3.4% last year.

Personally, I hope nuclear does triple in capacity. I have little doubt that wind and solar will dominate the energy system of 2050 but having more nuclear on the system can only help bring net zero closer. I know the naysayers will argue it's too expensive, takes too long and brings other externalities (particularly when it comes to decommissioning). Equally, the sector itself needs to demonstrate that it can deliver. The above chart on project delivery, cannot be repeated for Sizewell C and other Giga Watt scale projects. And those developing new SMR / AMR technologies need to demonstrate delivery, rather than promises. And of course none of this will be achieved without Governments setting clear and consistent regulations and policies.

Watch this space!

PS - for anyone interested in fusion (they don't like the nuclear word alongside), please refer back to F is for Fusion Energy

Further reading

As always, some further reading from the Energy Institute's New Energy World Senior Editor, Will Dalrymple!

News

Abu Dhabi nuclear power plant begins full commercial operation | Article Page

Good news for UK nuclear industry as government lays out billions | Article Page

Poland’s first nuclear plant moves forward with US agreement | Article Page

UK government plans to overhaul nuclear power plant planning rules in a year when global nuclear generation is set to hit an all-time high | Article Page

Features

Small is beautiful: UK plans for small modular nuclear reactors | Article Page

How civil nuclear power is being rehabilitated | Article Page

The evolution of India’s nuclear sector: learning from global trends | Article Page

Eventually, Europe’s first new nuclear plant for 15 years opens in Finland | Article Page (from 2022)

Comment/opinion

The advantages of thorium-based nuclear energy | Article Page

Can the world wait for new nuclear technology? | Article Page