Nuclear energy updated overview of key areas
Alexandru Agachi, November 2022
Build 1,000 new state of the art nuclear power plants in the US and Europe, right now. We won’t, but we should
- Marc Andreessen
I have come full circle on nuclear energy and now see it as way more attractive than most other forms of generating energy
- Fred Wilson
I want to be super clear. You should not only not shut down the nuclear power plants, but you should also reopen the ones that have already shut down. Those are the fastest to produce energy. It is crazy to shut down nuclear power plants now
- Elon Musk
I wrote my dissertation on the global energy trade at LSE and then pursued my research masters in energy sciences at Kyoto University, focusing on the thorium nuclear fuel cycle and the environmental impact of nuclear energy. I was a research assistant at the Kyoto University Research Reactor Institute (KURRI). Fukushima took place during my first semester there.
By the third day after the first reactor meltdown — Units 1, 2, and 3 would eventually melt, leading to three hydrogen explosions — I drove north with a lab colleague from China, to volunteer to help in the affected area. There were traffic jams on the highways going north, with people coming from all over Japan to help at the site — typically teams of colleagues from a same company, dressed in their company uniforms and sharing a car. There were so many people trying to enter the affected area in the first days after the first explosion, while the incident was nowhere near contained (Unit 4 was also a big unknown during those days), that authorities were already turning volunteers away from the region. It was an incredible sight — highways clogged with traffic, car after car of people dressed in their company uniforms, coming from all over Japan and voluntarily going to help at the site of an active nuclear disaster.
This is an updated overview on nuclear energy that I did for myself in 2022 to review key areas of nuclear energy, in the wider energy context. I remain a solid believer in the potential of civilian nuclear energy, and a solid disbeliever in anyone trying to tackle the energy challenge without tackling baseload power.
Nuclear energy — a Navy brat
Like several of the most important innovations of the 20th century, nuclear energy was born out of military research in the US. For completeness, nuclear fission itself was discovered by German scientists Otto Hahn and Fritz Strassmann in 1938. They observed that 1. uranium nuclei had a tendency to fission into several particles when bombarded with neutrons, and 2. this reaction releases a significant amount of energy. However, nuclear research really took off in the 1942–1946 period in the form of the Manhattan project, led in large part at the Los Alamos laboratory in the US — where nuclear research continues to this day. This project was of course mandated and driven by the US military during the Second World War. It is important to remember that at that time nuclear bombs were seen by the US military as simply the latest arsenal/weapon at their disposal, and people certainly didn’t have the governance/legal/cultural context that nuclear weapons have acquired today (and which we are perhaps fast losing as we speak).
The birth of nuclear energy on the largest battlefield of the 20th century, meant that nuclear energy spent its first decade as a military application with potential civilian uses. This matters because funding, specifications, research — everything in the first decade or two, regarding nuclear energy, was driven by military considerations and priorities. And indeed it is again within the US military, the Navy more specifically, that nuclear energy was first harnessed for power generation purposes. Under the direction of Admiral Hyman Rickover, the first nuclear reactors were developed for the US Navy in the late 1940s.
We can point to two choices that plagued the civilian nuclear energy as a result of its military birth, as two examples among many:
- the first is the choice of water, as a moderator and coolant for nuclear reactors. Other interesting technologies were known already, such as sodium, which was actually explored by the Navy (and even more impressively liquid metal cooled fast reactors were equally studied — a design considered today, 70 years later, as “advanced”). But Rickover’s team chose water as Navy engineers were quite used to working with it. This technology, the light water reactor (LWR), was selected for submarine propulsion, and started hogging funding in the 1950s over all other competing technologies. Today still, virtually all reactors in the world use water, although since the 1940s we’ve known that other materials presented numerous benefits over water
- the second choice, which took place a bit later, was the abandonment of thorium as nuclear fuel, in favor of uranium. Here again, in spite of obvious civilian advantages already known at the time (much more abundant in the Earth’s crust, a much higher melting point…), thorium was discarded as it could not be used to produce fissile material for military uses — a bug for the military, but a huge feature for civilian industry then as now. And yet this resulted in the US ending thorium research, with mostly labs in France and Japan continuing this line of work, the latter until today still. And hence, at the military’s behest, the entire global energy industry was built on a fuel that had a lower melting point, was less abundant in the Earth’s crust, and can be used duplicitously to create military grade fissile material that can be used in bombs.
The reason to start here is because I think it is vital for anyone looking at the nuclear energy industry today to understand how sub optimal it is, due to military choices that went against civilian optimization. Had we prioritized the latter in the 1940s and 1950s, arguably the global nuclear energy industry would look very differently today.
Wait but what is nuclear power?
Nuclear power is founded upon a simple principle: there are certain isotopes, like uranium-235 (naturally present in the Earth’s crust), which, after capturing a neutron, become unstable, and are very likely to split into multiple smaller fragments. When this happens, they release large amounts of heat. You then use this heat to boil water for example, and with the resulting steam you move a turbine and produce electricity. A nuclear reactor has been compared to “a giant boiler” and it’s not a wrong description.
Something worth noting here is that nuclear energy is a very dense, or very concentrated, source of energy: one tonne of uranium can produce 44 million kWh of electricity, which is equivalent to 22.000 tonnes of coal or 8.5 million cubic meters of natural gas (Tsoukalas et al. 2014).
Contemporary terminology around nuclear reactors is extremely unhelpful:
In its Energy and Water Development and Related Agencies Appropriations Act of 2020, the US Congress defined an advanced reactor as “any light water or non-light-water fission reactor with significant improvements compared to the current generation of operational reactors.” They might as well have said “anything better is advanced…”
A further confusion arises from the fact that a lot of the technologies included in this “advanced” categorization, were actually developed several decades ago, in the US, France, Japan, the UK and other countries. Terrani of the Oak Ridge National Laboratory, correctly points out that “today’s ‘advanced reactors’ closely resemble their 1950s–1970s predecessors in: core configuration; materials in structure, core, and fuel; approach to [fuel] qualification; and control systems” (Terrani 2019 quoted in Lyman, 2021).
Now if you consider yourself an innovator and want to skip over these old concepts since they doesn’t sound very “advanced,” please remember that in nuclear energy, like in finance and other critical fields, old and boring can be good news. First, while they are several decades old, these advanced designs represent several step changes from the reactors currently deployed around the world. Second, this means they have been researched, and for some of them tested, for decades instead of being the latest concoction in a garage. In short, iPhone in a garage is cool; nuclear reactor in a garage is not always cool.
Another terminology widely used but not particularly helpful either is based on generations of reactors. This widely spread definition goes like this:
Generation I: the early nuclear reactor prototypes of the 1950s and 1960s
Generation II: the commercial version of these prototypes, which accounts for almost all installed nuclear capacity today. What this tells you however is that pretty much all nuclear power plants installed around the world today are based on 1960s technology, and were built in the late 60s, 70s and 80s. Now think of 1960s cars and phones: this is just how old most nuclear power plants are in technological terms. It is vital to assess nuclear accidents like the one at Fukushima through this prism, understanding (beyond other errors pointed out in its aftermath) that this was a 1960s technology — the Fukushima Daiichi nuclear power plant was built in 1967 with reactors added until 1976.
Generation III/III+: while built on top of Generation II technology, Generation III/III+ reactors incorporate several additional decades of research, mostly in the areas of operational efficiency and safety. As a group these reactors tend to operate at higher temperatures (= higher thermal efficiency) and feature an array of new safety features, active and passive, that current reactors do not have. The first such Generation III reactors were recently built in China (Taishan, 2 reactors) France (Flamanville, 1 reactor), Finland (Oikuloko, 1 reactor) and the UK (Hinkley Point, 2 reactors).
Generation IV: This is where we fall again on the US Congress’s open ended definition of “everything better is advanced.” Pretty much everything that is significantly different/better than Generation III, falls under this Generation IV definition. If I think about the most important features that come up in this differentiation, Generation IV reactors tend to display one or several of the following:
· the use of moderator and coolant substances other than water;
· operating on a fast, instead of thermal, spectrum,
· the use of fuels other than uranium or uranium-plutonium mixes;
· much smaller size and power;
· widespread passive safety features.
In sum, Generation IV reactors tend to be more efficient, to service several industries (electricity but also industrial heat and desalination for example), to generate less waste, to possibly service smaller needs (remote towns, mining communities, military bases), and to be a whole lot safer.
As this terminology is so vague and unhelpful, I think for now we have to continue assessing reactor designs by paying attention ourselves to their most important technological parameters.
For me these are:
· Spectrum
· Fissionable material
· Cycle
· Moderator
· Coolant
· Capacity
· Safety features
Everything you look at in my opinion can be meaningfully classified along these parameters.
Lyman, Edwin. Union of Concerned Scientists, 2021, Nuclear Power: Present and Future / “Advanced” Isn’t Always Better, https://www.jstor.org/stable/resrep32883.6.
TSOUKALAS, LEFTERI H., et al. “A Future Role for Nuclear Energy?” Understanding the Global Energy Crisis, Purdue University Press, West Lafayette, IN, 2014.
The politics of nuclear energy
Politics will always have a tremendous role to play in capital intensive, heavily regulated, industries such as nuclear energy. It convenes though to have a good handle on the current political climate in key countries insofar as nuclear energy is concerned.
United States:
The last four US administrations, including the current Biden one, were each vocally supportive of domestic nuclear energy and its expansion. These are a few noteworthy legal and regulatory developments over the past decade or two in the US:
In 2001, under leadership from the US Department of Energy (DoE), the Generation IV International Forum (GIF) was established as an international cooperation body with the goal of making Generation IV reactors available for commercial deployment by 2030. So far more than 20 countries have participated in this forum.
Source: https://www.gen-4.org/gif/
In 2005 the Price-Anderson Act of 1957 was revised again, capping liability for a nuclear plant owner to 12.6 billion USD in case of an accident (an industry wide insurance scheme is furthermore in effect regarding this amount).
Source: https://www.cbo.gov/sites/default/files/112th-congress-2011-2012/reports/08-03-nuclearloans.pdf
In 2008, the US Congress approved US$18.5 billion in loan guarantees towards the domestic construction of nuclear power plants, and President Obama recommended tripling this amount to over 54 billion. However it is worth noting that over 122 billion had been requested by the nuclear industry (Squassoni, 2012).
The DoE further committed to sharing design and licensing costs for first of a kind reactors (estimated US$ 281 million). In addition, delay insurance was made available to the first 6 new reactors built: 500 million for each of the first two, and 250 million to each of the following four.
In 2011, the Nuclear Regulatory Commission (NRC) approved for the first time a Generation III reactor design, the Westinghouse AP1000. To put this event in context and explain why it is such a big deal for nuclear energy companies:
- NRC guidance for applications is 4500 pages long
- a typical application can easily exceed 12000 pages of content
- the approval process can take over a decade
- costs are estimated at anywhere from tens of millions (NIA estimate in 2021 https://www.nuclearinnovationalliance.org/unlocking-advanced-nuclear-innovation-role-fee-reform-and-public-investment) to half a billion dollars (NuScale estimate of its own approval process costs).
In 2019, the Nuclear Energy Innovation and Modernization Act (NEIMA, 2019) was passed:
This brought transparency to NRC fees, capped NRC annual fees for existing reactors, and most of all directed the NRC to develop a regulatory approval framework for the approval of new designs for nuclear reactors by 2027.
In 2020 we had the Energy Act of 2020:
This Act authorized US$ 6.6 billion worth of funding towards advanced nuclear reactors/energy over the following 5 years.
It also authorized and supported the development of a domestic High-Assay Low-Enriched Uranium supply chain (HALEU, used in many advanced nuclear reactor designs).
Breakdown of nuclear funding in this Act by functional area:
Source: Energy Act 2020 https://science.house.gov/imo/media/doc/Energy%20Act%20of%202020.pdf
In 2020, NuScale’s Small Modular Reactor, received NRC approval as a nuclear reactor design ready to be deployed commercially.
In 2021, The Infrastructure Investment and Jobs Act (also known as the Bipartisan Infrastructure Deal, 2021), earmarked USD 2.5 billion for the Department of Energy’s Advanced reactor demonstration program.
Source: https://www.world-nuclear-news.org/Articles/Nuclear-supporting-infrastructure-bill-becomes-US
In 2022, Oklo, a promising startup, saw its reactor design approval application rejected by the NRC. Analysts noted however that the application was 600 pages long, while the successful application of a “rival” startup, which was based on the same standard technology as the existing commercial reactors, was over 12000 pages long.
China:
All Chinese administrations of the past few decades have been supportive of the development of domestic nuclear energy. Jiang Mianheng, the former President Jiang Zemin’s son, oversaw core parts of China’s nuclear energy approach for a long time. And in particular, in 2015, President Xi re emphasized the advancement of nuclear energy as a key priority for China through a new industrial strategy policy for the nation, which came to be known as “Made in China 2025.” As part of this long term, progressive and consistent political position on nuclear energy it is difficult to point to one specific recent legislative highlight, but suffice to say that the building of nuclear plants in China is humming along.
Source: http://www.cittadellascienza.it/cina/wp-content/uploads/2017/02/IoT-ONE-Made-in-China-2025.pdf
European Union:
The European Union has become perhaps the world’s eminent regulatory power. Everyone took note therefore when in 2022, the European Commission adopted the Complementary Climate Delegated Act (CCDA). Following a detailed assessment, it concluded that “the analysis did not reveal any science-based evidence that nuclear energy does more harm to human health or to the environment than other electricity production technologies already included in the Taxonomy as activities supporting climate change mitigation.” It therefore classified nuclear energy as a transitional environmentally sustainable activity in the EU taxonomy (with specific rules).
This is by far the most important development in the EU at this level, effectively giving nuclear energy a green/sustainable label from a body whose taxonomies and recommendations have become inputs to similar processes around the world, be it in energy or in finance (this opened the door to sustainable finance financing nuclear power and indeed EDF, the French national electricity utility, has recently launched the first green bonds underpinned by nuclear power).
Also noteworthy, French President Emanuel Macron recently announced 1 billion euros in government support for the development of small modular nuclear reactors, with the French CEA also increasingly making its knowledge and support available to countries without nuclear industries around the world.
UK
UK politicians including Boris Johnson’s administration as recently as this year, have been vocal proponents of the expansion of domestic nuclear energy in the UK. In the Nuclear Energy (Financing) Bill, 2022, the UK introduced a regulated asset base (RAB) financing model, an extremely sound mechanism for promoting nuclear power in the UK, borrowed from what the UK government already does with respect to water, gas and electricity. Effectively this is a model governing in detail how construction and operation costs get split between investors/owners and consumers.
And as part of its British Energy Security Strategy (2022), the British government also launched The Future Nuclear Enabling Fund (FNEF) in order to support (mature) nuclear energy development projects in the UK, with GBP 120 million in initial funding authorized.
Sources: https://bills.parliament.uk/bills/3057
Current administrations are supportive of the expansion of domestic nuclear energy in key countries including the US, France, China and the UK. Other countries with supportive administrations are Japan and South Korea, as well as Russia and India. Overall nuclear energy today benefits from political support in countries representing the vast majority of the existing nuclear power park, as well as the majority of building activity.
Squassoni, Sharon. Federation of American Scientists, 2012, The Future of Nuclear Power in the United States / NUCLEAR POWER IN THE GLOBAL ENERGY PORTFOLIO, https://www.jstor.org/stable/resrep18936.6.
What about public opinion?
One can safely assert that the nuclear industry has done a horrendous job at marketing itself. The Netflix special on Tchernobyl is a typical “advertisement” for the nuclear industry over the past few decades. In particular when compared to other renewable sources of energy, whose claims of superiority can be easily dismantled yet are widely accepted by the public. And for such an old, regulated, and technically complex industry, perhaps it was to be expected.
What matters today nonetheless is that public opinion is a full, key stakeholder in nuclear energy. And public attitudes towards nuclear energy seem to ebb and flow with nuclear accidents — in particular the 3 that come to mind in the history of the industry. For nuclear energy to thrive in a country, it is probably sufficient for the population to not be drastically opposed to it. And when looking at public opinion surveys, it becomes evident that very few countries on this planet have a strongly anti nuclear public. Even in countries phasing out nuclear energy at the moment, a detailed look would indicate that beyond a small and committed minority, the population is hardly dogmatic and consistent about this.
For example a study on European attitudes concluded to a significant increase in support for nuclear energy in 17 out of 27 countries (and a decrease in only 2 countries) — and this study was done before the Ukraine war and the explosion in domestic energy prices in Europe. It further pointed out that a country’s experience with nuclear energy seems to be the strongest determinant in its population’s attitude towards it, with countries with large nuclear parks being significantly more in favor of nuclear energy than countries without nuclear power plants. The study further concluded that the European public highlighted energy diversification (64%), a decrease in dependence on oil (63%), and a reduction in GHG emissions (62%) as core benefits of nuclear energy for them1.
Across public opinion polls, a majority of Americans is seen as supporting the expansion of their nuclear industry2 (51%, 2022), of British3 (65%, 202), of Australians4 (55%, 2021), of Polish5 (74%, 2021), of Japanese6 (53%, 2022), and of French7 (64%, 2021). Even in Germany support for nuclear power has significantly increased in light of the Ukraine war, with a majority of Germans favoring the continuation of the country’s remaining nuclear power plants as of 20228.
The nuclear energy industry has done a terrible job at education and marketing, in spite of the brilliant factual resources that are published by a few intergovernmental and industry bodies such as the NEA/NIA/WNA. Yet in spite of its best efforts at losing the public, it seems clear that public support for nuclear energy has consistently increased in many key countries over the past few years. The Ukraine war in particular, mixed with unsavory geopolitics and sky high energy prices, has galvanized support for nuclear energy in many Western countries.
1: https://www.iaea.org/sites/default/files/50104703435.pdf
2: https://news.gallup.com/poll/392831/americans-divided-nuclear-energy.aspx
4: https://www.afr.com/companies/energy/half-of-left-wing-voters-support-nuclear-power-20210326-p57eci
5: https://www.gov.pl/web/klimat/74-polakow-popiera-budowe-elektrowni-jadrowych-w-polsce
6: https://www.japantimes.co.jp/news/2022/03/28/national/nuke-power-poll/
7: https://www.world-nuclear-news.org/Articles/Survey-shows-growing-public-support-for-nuclear-in
8: https://www.world-nuclear-news.org/Articles/Wide-public-support-for-keeping-German-reactors-on
The global nuclear power park
There are 440 nuclear reactors in operation today, operated across 32 countries. This represents 10.2% of global electricity production, and almost 30% of low carbon electricity produced across the world (WNA, 2022).
In the United States, there are 93 commercial reactors operating today, making it the largest global producer of nuclear energy with about 30% of the global total. With an average age of 40 years, the US also has the oldest nuclear park in the world (EIA, 2022). While most US reactors were built in the 1970s, as of August 2021 roughly all of them had received a license renewal, with 6 of them receiving even a second license renewal (NRC, 2022). This effectively presages an extra 20–40 years in operation for most of them.
And to a large extent this is the blessing, and the curse, of nuclear energy — a country like France can build from scratch, over roughly two decades, a nuclear power park that provides 70–80% of its electricity and then it doesn’t need to build new plants for 60–80 years (expected life expectancy of the current nuclear park following an initial expectation of 40 years, with 2 license renewals of 20 years each, as it is currently happening in the US). And while R&D and innovation freely hums along in government and private laboratories, you cannot dismantle a complex and regulated construction industry for 60–80 years, and then expect to pick up with radically new designs exactly where you left off half a century earlier.
Also in the US, four new enrichment plants are under construction or awaiting approval to start construction (Squassoni et al. 2012).
In the European Union, between now and 2050, EUR 50 billion will be invested towards extending the life of the current nuclear park, and over EUR 400 billion towards its expansion (Thierry Breton quoted in RFI, 2022).
France in particular announced the construction of 14 Generation III reactors (Quattry et al., 2022).
And in the United Kingdom, Boris Johnson set the goal of increasing nuclear energy from 15% to 25% as a percentage of total electricity production.
There are currently 52 reactors under construction around the world, totaling 47–54 GW of capacity (depending on the source):
Source: World Nuclear Report 2021 as cited in Quattry et al. 2022
As we can see most new builds are planned in emerging markets and indeed, the IEA expects emerging markets to account for 90% of the global nuclear growth (2022).
In Asia, South Korea has a nuclear friendly administration under President Yoon Sukyeol, with the stated objective of increasing nuclear power to 30% of total electricity produced in the country.
Japan has recently (2022) announced its intent to restart its entire nuclear power park and of continuing to rely on nuclear energy for a large part of its domestic electricity needs.
But it is China really that is moving the needle, with 150 reactors planned in the next 15 years, more than the rest of the world combined currently.
While the world is currently building 5 new reactors per year, it is generally accepted by intergovernmental bodie that to meet the IPCC’s objectives, we will need to build a lot more reactors. In some estimates, nearly 600 GW of new capacity will need to be installed by 2050, or 500 large reactors in total, or 20 per year roughly (Lyman, 2021).
IEA, 2022, Nuclear Power and Secure Energy Transitions, https://www.iea.org/reports/nuclear-power-and-secure-energy-transitions.
“Nuclear Explained U.S. Nuclear Industry.” U.S. Nuclear Industry — U.S. Energy Information Administration (EIA), https://www.eia.gov/energyexplained/nuclear/us-nuclear-industry.php.
“Nuclear Power in the World Today.” Nuclear Power Today | Nuclear Energy — World Nuclear Association, WNA, https://world-nuclear.org/information-library/current-and-future-generation/nuclear-power-in-the-world-today.aspx.
Quattry, Steven, et al. 2022, The Nuclear Revival: Embracing a Clean, Reliable and Safe Source of Energy, https://www.morganstanley.com/im/en-us/individual-investor/insights/articles/the-nuclear-revival-embracing-a-clean-reliable-safe-source-of-energy.html.
Reactor License Renewal | Nrc.gov. NRC, https://www.nrc.gov/reactors/operating/licensing/renewal.html.
Rfi. “Europe Nuclear Plants ‘Need 500 Bn Euro Investment by 2050’: Eu Commissioner.” RFI, RFI, 9 Jan. 2022, https://www.rfi.fr/en/europe-nuclear-plants-need-500-bn-euro-investment-by-2050-eu-commissioner.
Nuclear garbage
What is nuclear waste? Effectively it is simply the physical waste generated by the operations of a nuclear reactor. Most of the time, when we talk about waste we talk about the nuclear fuel waste that is produced by a nuclear reactor. And what exactly is this fuel waste?
When bombarded with neutrons, the fission of the uranium-235 nucleus creates several smaller nuclei. These smaller nuclei are unstable. They will therefore undergo beta decay towards more stable nuclei. This decay releases beta or gamma radiation, in a reaction such as the following:
235U + n → 236U → 140Xe + 94Sr + 2n
Now it is important to note that most fission products created have extremely short half lives, such as Xenon-140 at 14 seconds and Strontium-94 at 75 seconds.
Other elements are called long lived fission products (LLFP) as they have longer half lives: Iodine-131 at 8 days, Cesium-137 at 30 years, and Strontium-90 also at 30 years. Almost all fission products have half lives of less than 90 years, with their radioactivity decreasing by a factor of 1000 after the first 40 years. As such, almost all radioactive waste is either extremely short lived or relatively easy to handle.
It is the actinides that pose the largest challenge: these are several synthetic radioactive chemical elements belonging to the Americium, Plutonium, and Curium families. Their half lives are thousands and for some of them millions of years. This said, we have to put this in context as the quantities produced are extremely small: one tonne of nuclear waste contains about 100 grams of Americium for example (a 1000 MWe nuclear reactor produces about 27 tonnes of spent fuel annually so about 2.7kg of Americum per year per reactor).
At a high level, 90% of the waste produced by a nuclear reactor is considered low level, 3% is considered high level, and only 1% is radioactive. To further place in context the low quantities of nuclear waste, it is estimated that all the nuclear fuel waste produced by the US nuclear energy industry (the largest one in the world), over the past 60 years, would “fit on a football field, stacked 20 feet high” (Goldstein and Qvist, 2019). It is further estimated that “an American’s entire lifetime of electricity generated by nuclear power would produce long-term waste that fits in a soda can.” (Quattry et al., 2022).
What happens to this nuclear waste? Once removed from the reactor, it is typically kept in a pond, under water, for at least a year. This helps with both (residual) heat removal and with radioactivity containment. Past this period of at least a year, the fuel is taken out and placed into gas filled, steel cylinders called casks by the industry. This is their long term storage form. While countries like the United States do not allow for fuel reprocessing, countries like France and the UK do. In this circular nuclear fuel cycle, the spent fuel is reprocessed in order to create new fuel from it (“MOX” or mixed oxide fuel). This significantly reduces both the amount of final fuel waste created, as well as the uranium needs of the local nuclear industries. Following this process, France, probably the world leader in this process, has reprocessed over 26.000 metric tons of heavy metal to date (a ton unit used in the nuclear industry to refer to uranium, plutonium, thorium and their mixes) (Hibbs, 2018). The United States, where this process is anachronistically forbidden, has accumulated about 60.000 metric tons of fuel waste to date, which sits in ponds and storage casks at 121 sites spread across 39 states, fault of the Yucca Mountain repository site still not being operational almost 25 years after its intended opening date (1998). Last, here is an overview of nuclear waste around the world today:
Source: Tsoukalas et al. (2014)
Goldstein, J S, and S A Qvist. “Only Nuclear Energy Can Save the Planet.” The Wall Street Journal, Dow Jones & Company, 11 Jan. 2019, https://www.wsj.com/articles/only-nuclear-energy-can-save-the-planet-11547225861?ns=prod%2Faccounts-wsj.
Hibbs, Mark. Carnegie Endowment for International Peace, 2018, THE FUTURE OF NUCLEAR POWER IN CHINA / CHINA’S CHOICE FOR NUCLEAR POWER AND A CLOSED NUCLEAR FUEL CYCLE, https://www.jstor.org/stable/resrep26911.8. Accessed 2022.
TSOUKALAS, LEFTERI H., et al. “A Future Role for Nuclear Energy?” Understanding the Global Energy Crisis, Purdue University Press, West Lafayette, IN, 2014.
How dangerous is civilian nuclear energy?
Just like there is no perfectly renewable source of energy and all sources of energy are on a spectrum of sustainability and environmental impact, so it is with safety. Accidents and unintended byproducts unfortunately do happen for all sources of energy. All we can do is rationally and factually assess all sources of energy and place them on a spectrum in terms of the human deaths they are accountable for. And we need to optimize for this, by expanding the safer ones and reducing the less safe ones.
Radioactivity is everywhere around us the same way sunlight or wind are everywhere around us. Our planet naturally emits radioactivity and we grow with it everywhere around us. An entire branch of medicine harnesses radioactivity to treat cancer and save lives in hospitals large and small, all around the world. My former nuclear energy professor, a well respected physicist in Japan, used to often say that he would much rather drink a glass of water with a bit of cesium (cesium is radioactive) in it, than drink the Latte Ventis I used to come to the lab with. Now I never switched to the Venti Cesium latter, but he had a point that we shouldn’t irrationally fear radioactive substances hidden under a mountain 1000km away from our cities (nuclear waste), while we breathe and ingest pollutants and contaminants away on a daily basis in our regular lives. And indeed if you want to live a healthier or safer or longer live, protesting in front of the neighborhood nuclear power plant is, objectively, not the right place to start your journey.
The common method to assess energy safety is to measure the number of deaths, including from accidents and unintended consequences like air pollution, per unit of energy produced from a specific source. It is easy and quick to realize that on this objective metric, nuclear energy is an extremely safe source of energy, in spite of the 3 nuclear accidents that took place over the past 50–60 years around the world and that we all know of. As Quattry et al. (2022) point out, “Arguably the biggest misconception about nuclear power is safety.” Indeed, Ritchie and Roser (2020) calculated that deaths from nuclear energy are 99% lower than from coal and 97% lower than from natural gas. At 0.07 deaths / TWh, nuclear energy is very much comparable to solar (0.02) and to wind (0.04). Below is another estimate which actually places nuclear energy lower than wind, and right between wind and solar below.
Source: Ritchie&Roser (2020) at https://ourworldindata.org/nuclear-energy#:~:text=We%20 see%20massive%20differences%20in,hydropower%20are%20more%20 safe%20yet
The same researchers point out a year earlier (2019) that while the accident at Chernobyl caused the premature deaths of 4000 persons, as much as 4.5 million people die every year from the consequences of air pollution stemming from fossil fuels (Source: Ritchie&Roser at https://ourworldindata.org/outdoor-air- pollution?country=#citation).
The nuclear industry is of course working towards further improving its safety record. Most existing commercial reactors in the US are designed for one in one hundred thousand year damage compliance. The general objective of next generation reactors is to push this to one in ten million years (Tsoukalas et al., 2014). And looking to the future, Pushker and Kharecha (2013) concluded that “nuclear power could additionally prevent an average of 420 000–7.04 million deaths and 80–240 GtCO2-eq emissions due to fossil fuels by midcentury, depending on which fuel it replaces.”
We see here that wind, solar, and hydropower are not perfectly safe sources of energy, the same way they are not perfectly renewable sources of energy. Nuclear energy is very much comparable to solar and wind in terms of safety track record, including once we account for any and all accidents that occurred over its lifetime. Nuclear energy also appears as the only safe source of baseload power that we have today for most of the planet.
Kharecha, Pushker A., and James E. Hansen. “Prevented Mortality and Greenhouse Gas Emissions from Historical and Projected Nuclear Power.” Environmental Science & Technology, vol. 47, no. 9, 2013, pp. 4889–4895., https://doi.org/10.1021/es3051197.
TSOUKALAS, LEFTERI H., et al. “A Future Role for Nuclear Energy?” Understanding the Global Energy Crisis, Purdue University Press, West Lafayette, IN, 2014.
Fundamental economics of nuclear power
The economics of nuclear power are one of the most contentious points in the wider energy debate. They are also, with public opinion, the main if not the only, barrier to the expansion of the nuclear industry — I think of upfront capital costs and public opinion as the two levers that govern the nuclear industry over time and across different countries.
Energy costs are usually expressed in “megawatt-hour”: a megawatt-hour is simply 1 megawatt, or 1000 kilowatts, used continuously for one hour. This is also what your electricity meter at your home indicates, and how your electricity provider charges you — you’ve certainly taken such MWh/kWh meter readings over time.
We can think of the economics of nuclear energy as the addition of several cost components, including 1. capital costs, 2. operating and maintenance costs (O&M), 3. external costs and 4. other costs.
1. Capital costs, including financing and interest
Overnight cost = capital cost exclusive of financing costs
Financing costs = interest expenses on the (large) debt burden of nuclear power plants, which by definition also reflects any cost overruns and delays in construction
Construction time regarding duration in the financing equation = time from first pour of concrete to connection to electricity grid
Capital costs = overnight costs + financing costs
While this is the generally accepted terminology, you will find that authors use these terms somewhat loosely. However any serious study will be very clear on what is included in each of these terms, and, knowing these terms as well as the most important parts of the total costs, you should easily be able to find yourself around and know what numbers you’re reading.
2. Operating and maintenance costs (including provisions for decommissioning and waste)
While critics of nuclear energy often like to point out to the “hidden” costs of radioactive waste and decommissioning, it is important to highlight that we have over 60 years of hindsight on the global nuclear energy industry, and these costs are always factored in and accounted for — in all serious/mainstream studies, they are neither “hidden” nor “opaque.”
3. External costs
These can be called externalities as well.
4. Other costs, such as system costs and taxes
Here we have an additional indirect set of externalities, which the nuclear industry correctly argues should be imposed on all sources of energy and in particular on intermittent sources of energy such as solar and wind.
Now let’s dive in a bit deeper into Capital, O&M, and External costs.
1. Capital costs
Existing nuclear power plants are large scale infrastructure projects, recently plagued by the cost overruns and delays typical of megaprojects around the world. As such, capital costs are a much more significant part of the total costs of nuclear power generation than for any other source of energy. The below aims to indicate roughly what these costs are and how they are split. This is of interest in particular given the community’s focus on lowering the capital costs of nuclear power plants nowadays — understanding the historical breakdown of these costs is key to assessing such claims.
Source: https://world-nuclear.org/information-library/economic-aspects/economics-of-nuclear-power.aspx
Source: https://www.innovationreform.org/wp-content/uploads/2018/01/Advanced-Nuclear-Reactors-Cost-Study.pdf
2. Operating and maintenance costs:
Fuel costs:
A key characteristic of nuclear energy is the very high concentration of its fuel, uranium. This means that very small quantities of uranium are needed to generate very large amounts of energy: one kg of uranium fuel generates 44.000 kWh of electricity, equivalent to 22 tonnes of coal, or 8500 cubic meters of natural gas. And such a kg of uranium fuel, UO2, costs approximately US$1663/kg (https://world-nuclear.org/information-library/economic-aspects/economics-of-nuclear-power.aspx), broken down per the following:
This US$1663/kg amounts to 0.46 cents USD /kWh. This also means nuclear energy is very insensitive to changes in the uranium price. NEA estimated that the nuclear power industry would barely be affected by a 50% change in the price of uranium, while coal and gas are highly sensitive to commodity prices of course.
Costs of dealing with radiactive waste are also small: for example the US$26 billion used fuel programme is funded by a 0.1 cents USD / kWh charge.
Other operating costs are of course related to the ongoing operation of a nuclear power plant, which include labor. More on such O&M costs can be found in the WNA report on the costs of nuclear power cited above.
An interesting point here however is that many innovative reactor designs would limit, or eliminate altogether, onsite labor, which could be interesting from an economics point of view. These claims are speculative for now however, and in any case they pale compared to the capital costs above, which should remain the main focus for anyone trying to increase the competitiveness of nuclear energy.
Overall, O&M costs account for 66% of the total operating cost.
And last, decommissioning costs account for 9–15% of the initial capital cost of a nuclear power plant, which represents about 5% of the cost of electricity they produce.
Source: https://world-nuclear.org/information-library/economic-aspects/economics-of-nuclear-power.aspx
3. External costs:
Externalities often come up when discussing nuclear energy, and for good reason. Several studies have tried to quantify, and for some of them compare, the externalities produced by nuclear energy. One of the first such attempts was perhaps ExternE, a collaboration between the European Commission and the US Department of Energy. While it took into account emissions, dispersion, and even radiological impact from uranium mining, it did not quantify the effects of global warming, meaning the figures below would be significantly larger for fossil fuels if these were accounted for (in USD cents per kWh):
Source: WNA Economics of Nuclear Power, 2021
A more recent (2014) study undertaken for the European Union, by Ecofys, concluded to total external costs of nuclear energy of EUR18–22 / MWh, broken down into EUR5 for health impacts, EUR4 for accidents, and EUR12 for resource depletion (Grave et al. 2016).
Regarding nuclear accidents, which are often pointed out as a “hidden” cost of nuclear energy, these have also been studied and broken down openly. While Grave et al. (2016) estimate them at EUR4 / MWh, another study for the European Commission, by D’haeseleer (2013) from the University of Leuven, put them in the 0.3–3EUR/MWh range.
When it comes to externalities, nothing comes close to the cost of climate change caused by fossil fuels. Nuclear energy comes out of an externality analysis somewhat neutral in the grand scheme of things (compared to all other alternatives), while fossil fuels are heavily penalized.
Overall, here is a breakdown of the costs we have just seen for a nuclear power plant:
And what this looks like concretely in a recent Generation III EPR build, in the UK:
What jumps at us of course is the large interest charge, which is due to the high capital costs of nuclear power plants. This is why typically you will see analyses done with several discount rates, usually 3%, 7% and 10%, and why, which one of these rates you focus on will drastically impact the competitiveness of nuclear power. For example a study by the IEA and NEA (2020) concluded that “At a 3% discount rate, nuclear is the lowest cost option for all countries. However, consistent with the fact that nuclear technologies are capital intensive relative to natural gas or coal, the cost of nuclear rises relatively quickly as the discount rate is raised. As a result, at a 7% discount rate the median value of nuclear is close to the median value for coal [but lower than the gas in CCGTs].”
Nuclear energy and total generating costs:
What are usually called “generating costs”: in this measure we typically find included the operational costs, fuel, and capital used strictly to operate (and not build for example) the respective power plant, irrespective of which type of plant it is (gas, nuclear, coal…). This measure therefore highly favors capital intensive industries where very large amounts of capital need to be deployed upfront at onset, and downstream at decommissioning. Most of all, this favors nuclear power, the highest capital cost power generation industry, because here you assume you simply own a nuclear power plant (gifted by a relative) and you are looking at what it costs to produce energy from this plant going forward.
On the basis of this metric, nuclear power is not just competitive, but downright cheap and decreasing. In the US, total generating cost for nuclear energy was US$ 29.37/MWh in 2020, 4.6% lower than in 2019, and 35% below 2012.
Source: NEI (2021)
The reason we touch on total generating costs here is the active debate in several countries on whether to stop nuclear energy, or to extend the lifespan of the already existing nuclear power plants. While this total generating cost is not a very fair comparison tool for the reasons explained, it is useful in this one context — countries considering whether to prematurely end the life of a nuclear power plant. In this setup, construction costs are a sunk cost by definition, and decommissioning costs can be assumed to be the same whenever one terminates the power plant. What we therefore see is that such countries turn off a perfectly cost competitive source of energy — probably the cleanest and cheapest source of energy they have actually. And indeed, the IEA (2022) notes that “Lifetime extensions are a very cost-effective source of low emissions electricity … Nuclear plays a complementary role, contributing to system stability, expanding the suite of low emissions sources, and stepping up where renewables are constrained.” They add that the capital cost for nuclear power plant extensions by 2030 will be US$500–1100 per kW, leading to an LCoE below US$40/MWh, which is cost competitive against any other existing source of energy.
Nuclear energy and Levelized Cost of Energy (LCoE)
The appropriate way to compare different energy sources of course is to focus on the entire life cycle required to generate power from that electricity source: construction costs, fuel, operations and maintenance costs, and decommissioning. This is why we usually don’t focus on the “total generating” or “operational” costs above, but rather on a holistic metric called the Levelized Cost of Energy (LCoE). The LCoE is the total cost to build, operate, and decommission a power plant over its lifetime, divided by the expected total output over that lifetime, measured typically in megawatt hour.
One of the most authoritative sources on the LCoE of different energy sources is the EIA’s annual report. As such, here is how different sources of energy compare in this report:
Source: https://www.eia.gov/outlooks/aeo/pdf/electricity_generation.pdf
https://www.eia.gov/outlooks/aeo/assumptions/pdf/table_8.2.pdf
And here is another comparative study, by the IEA this time:
Nuclear LTO: Long term operation of nuclear power plants, effectively takes into account extensive refurbishment costs leading to an extension of the life expectancy of installed nuclear capacity.
Source: https://www.iea.org/reports/projected-costs-of-generating-electricity-2020
We see that nuclear energy clearly appears as a competitive source of energy on this LCoE basis.
But nuclear power costs can also vary considerably from country to country. Jan Emblemsvag (2021) provides such a comparison, also illustrating different discount rates and their effects:
*LCoE for plants built in 2015–2020
Source: Data and graph taken from the report by Columbia Threadneedle (2022).
Another comparison with slightly different figures is provided by the World Nuclear Association (2021):
Note: the costs here were essentially all costs except for interest during the construction period.
Source: https://world-nuclear.org/information-library/economic-aspects/economics-of-nuclear-power.aspx
These differences between countries have mostly been explained through differences in:
· local labour costs;
· more recent experience building reactors;
· streamlined nuclear power plant designs (i.e. South Korea);
· economies of scale from favoring the building of multiple units;
· streamlined licensing
Last, an important fact in the energy industry is how the costs of building nuclear power plants, instead of decreasing, have vastly increased between the 1960s and the 2000s — before stabilizing recently, and even decreasing again in some countries. I gathered here some such historical comparisons as available across the literature and indeed we see a literal explosion in costs across countries. As an example, the cost of building 6 Generation III EPR reactors in France today is estimated at 56 billion euros. For comparison, the entire existing French nuclear power park, all 56 reactors and 63GW of energy capacity, cost 85 billion euros to build in 2010 prices (WNA 2021), representing an extremely competitive cost of EUR1335 / kW (versus EUR 5700/kW for the six EPR reactors).
*NEA Projected Costs of Generating Electricity, 2020
**WNA Economics of Nuclear Power, 2021
***EIA Capital Cost and Performance Characteristic Estimates for Utility Scale Electric Power Generating Technologies, 2020
****Cour des Comptes, 2012, EUR
A few key takeaways from a historical review seem to be that:
- we are obviously capable of building cost effectively nuclear power plants in a manner that is competitive with any other source of energy
- economies of scale are crucial in the nuclear industry. Economies of scale are achieved through few designs replicated across many sites similar in build, with multiple reactors per site (hosting two reactors per site led to a 15% decrease in cost per kW in China compared to a single reactor (WNA, 2021) and in the US, the difference in cost per kW between single and multi reactor existing plants is of 31.8% lower for the latter (NEI, 2021)).
- the nuclear energy industry has been (over)burdened with a very onerous regulatory regime, coupled with inconsistent policies. One can see attempts to recognize and reform this in the US’s recent efforts to modify the NRC approval and oversight process
- recent experience and skillsets are crucial. Countries like the US and France, in spite of hosting more reactors than any other country, most likely lag in terms of building capabilities countries like China, something reflected in the Vogtle (US) and EPR projects of late — it is not just public authorities but also companies that are rusty in the West, not having done much domestic building since the 1970s. What we see, similar to other complex domains (like rare earths mining in the US for example), is that we can very much lose domestic capabilities and experience if we do not purposefully maintain them
Positive news here are that:
· The IEA concluded that costs for building new nuclear plants have peaked (WNA, 2021)
· Several countries have shown stabilized new build costs (Lovering et al. 2016)
· Some, like South Korea, have even shown decreasing costs for new builds (Lovering et al. 2016)
· While the first Generation III builds (Vogtle in the US, the 3 sites in Europe) have all exhibited very large cost overruns and delays, we can reasonably expect the cost per reactor to significantly go down as these technologies go from FOAK to being replicated across many sites
Please note that typically LCoE metrics do not include interest charges, and we saw how high those can be for nuclear energy. LCoE metrics do include the capacity factor of different sources of energy however.
Given that nuclear energy is often benchmarked with other renewable sources such as solar and wind, when it comes to costs, the debates on renewable energy typically miss out several key points, which should be fully accounted for:
1. Renewable energy is not renewable. Everything is on a scale of sustainability, and nothing we currently have is perfectly renewable. The IAE for example points out that both wind turbines and solar panels depend on neodymium, dysprosium, terbium, europium, and yttrium, and that shortages of these materials are already forecasted and will impact their availability and their cost significantly (IEA 2021 cited in Rogner et al. 2021).
2. Solar and wind are intermittent sources of energy. As such it seems correct to factor into their LCoE an additional cost, namely the LCoS: the cost of storage.
It is generally accepted that as the share of wind and solar power in our total energy mix increases, significant new problems will arise. A study between the DoE NREL and the DoE EERE by Denholm et al. (2021) concluded for example that the difficulties and costs of the wind+solar mix will considerably increase. They point in particular to the gargantuan task of handling seasonal variations in conditions, in countries like Germany and Switzerland. It is estimated that Germany would need to install >1000GWh of energy storage just to handle the daily variations in wind and sun. The much tinier Switzerland, would require >3000 GWh to handle its extreme seasonal variations (Aegerter, 2017). The authors compare this with the largest battery storage system we currently have, Tesla’s system in South Australia, which has a capacity of all of… 0.2GWh (and in terms of sustainability keep in mind that batteries and storage systems are not built from air and water).
3. Building on the previous point, multiple reports highlight that due to their intermittent nature and distant location, wind and solar impose significant “system costs” on existing grids, which are borne by consumers, but not added to the costs of wind and solar power
The NEA for example points out in 2019 that integrating intermittent sources of energy into existing electricity grids will pose very large challenges and with very large associated costs that are currently not reflected / priced into any electricity market, which they estimate at US$8–50/MWh, versus US$1–3/MWh for nuclear energy. The IEA (2022) similarly highlights that reaching net zero by 2050 without accelerating nuclear expansion will cost an extra US$500 billion on renewables, storage and carbon capture, increasing the annual electricity bills by 20 billion by 2050. Building nuclear power plants seems expensive, but so is not building them, and we have to be honest with these costs across the aisle.
4. Subsidies
We often forget that wind and solar power are booming partly due to the immense subsidies they receive in many countries, whether in the form of direct subsidies or of indirect ones such as tax abatements. A simple example given by Quattry et al. (2018) is that in 2018 the nuclear industry received US$13 million in federal tax incentives in the US, compated to US$2 billion for wind, and 3.3 billion for solar, or 250 times more than nuclear. A study undertaken by Lazard (2021) similarly highlights that the competitiveness of swind and solar power remain extremely sensitive to subsidies.
Different studies increasingly try to account for some if not all of these costs. This said, these costs are usually not all factored into different studies, meaning that most studies portray an overly optimistic picture of the current costs of solar and wind power, and especially of the future costs of running the majority if not entirety of our national grids on these two renewables.
As economics is a major roadblock to nuclear power expansion, the nuclear world is, without surprises, focused on reducing costs as much as possible. Governments (i.e. the U.S. Congress), utilities, and nuclear energy companies, are all working towards reducing the costs of new nuclear builds. With a focus on innovation, an authoritative peer reviewed study was conducted by the Energy Innovation Reform Project (EIRP) in 2017. EIRP reviewed, with a transparent methodology, the expected costs of 8 new reactor designs currently in development:
Averaging these 8 advanced reactor designs, they arrived at the following expected future capital costs, operational costs, and LCoE:
Source: https://www.innovationreform.org/wp-content/uploads/2018/01/Advanced-Nuclear-Reactors-Cost-Study.pdf
These costs are highlighted as competitive with natural gas, whose LCoE is cited as US$42–78 / MWh, compared to the above range of 36–90.
The NEA’s forecasts generally move in the same direction, estimating that for new builds, we will go from a US$5000/kWe in 2020 to 4000 in 2025–2030. The reasons advanced by the NEA include maturing designs, recent experience, and lessons learned during FOAK builds in the 2010s.
While in light of past experience we have to take these forecasts with a grain of salt, it is interesting to see where the report authors and these 8 startups, think economies will stem from. At a high level, simpler designs, standardized designs, modular designs, factory based manufacturing (instead of onsite), parallel construction (instead of onsite sequential), are the main upstream points highlighted. These are naturally expected to lead to shorter construction times and delays. Once operational, a higher power density, higher efficiency, and to some extent less labor required for O&M, are the main areas of focus.
When it comes to future economics, there is a particular focus on the radically different, small modular reactors. Authors like Ingersoll et al. (2020) concluded that SMRs would achieve a capital cost as low as US$3000/kW. The small size of an SMR and its modular simplicity are usually put forward as key reasons for its expected lower capital costs. This is how economies are expected from SMRs:
Source: EFWG (2018) cited in Rogner et al. (2021)
This said, other authors reach the opposite conclusion however. They point out that specifically because of its small size, while construction costs would indeed be 40% less than that of that of a large reactor, the output would also only be 20% of that of a large reactor, leading to twice the cost per MW of capacity. And indeed so far, in the upstream stages, SMRs have been burdened with a similar experience as their larger peers: it is estimated that the NuScale SMR design had cost US$ 957 million to develop until March 2020, and that another 500 to 700 million would be needed in regulatory costs before commercial construction could potentially begin.
The main take aways would be that in terms of operating cost (i.e. existing nuclear power plants), nuclear energy is extremely cost competitive with the cheapest sources of energy.
Once we account for all costs, at an LCoE level, nuclear energy remains competitive with both fossil fuels (with a carbon taxation in place and/or carbon capture and storage but before even accounting for the actual costs of climate change), and with other renewables, for which a host of costs usually not reflected, from storage to system costs, need to be taken into account.
But the high upfront capital costs of nuclear energy remain a major barrier to the industry’s expansion. While tremendous effort is being applied across the political, regulatory, technological and construction areas to alleviate this, expectations of cost savings will need to be proven on the ground for both additional Generation III builds, and for advanced reactor designs (Generation IV).
Aegerter DS (2017). 100% Wind and Solar? A Seductive Myth. The Stern Institute Periodical Vol. 16
Denholm P, Arent DJ, Baldwin SF, Bilello DE, Brinkman GL, Cochran JM, Cole WJ, Frew B, Gevorgian V, Heeter J, Hodge B-MS, Kroposki B, Mai T, O’Malley MJ, Palmintier B, Steinberg D and Zhang Y (2021). The challenges of achieving a 100% renewable electricity system in the United States. Joule https://doi.org/10.1016/j.joule.2021.03.028
D’haeseleer, William D. European Commission, 2013, Synthesis on the Economics of Nuclear Energy, https://www.mech.kuleuven.be/en/tme/research/energy_environment/Pdf/wpen2013-14.pdf. Accessed Oct. 2022.
Grave, Katharina, et al. Ecofys & European Commission, 2016, Prices and Costs of EU Energy, https://www.isi.fraunhofer.de/content/dam/isi/dokumente/ccx/2016/report_ecofys2016.pdf. Accessed Oct. 2022.
International Energy Agency and Nuclear Energy Agency , 2020, Projected Costs of Generating Electricity, https://www.oecd-nea.org/upload/docs/application/pdf/2020-12/egc-2020_2020-12-09_18-26-46_781.pdf. Accessed Oct. 2022.
International Energy Agency (IEA), 2022, Nuclear Power and Secure Energy Transitions, https://iea.blob.core.windows.net/assets/016228e1-42bd-4ca7-bad9-a227c4a40b04/NuclearPowerandSecureEnergyTransitions.pdf. Accessed Oct. 2022.
Lazard, 2021, LAZARD’S LEVELIZED COST OF ENERGY ANALYSIS — VERSION 15.0, https://www.lazard.com/media/451881/lazards-levelized-cost-of-energy-version-150-vf.pdf. Accessed Oct. 2022.
Lovering, Jessica R., et al. “Historical Construction Costs of Global Nuclear Power Reactors.” Energy Policy, vol. 91, 2016, pp. 371–382., https://doi.org/10.1016/j.enpol.2016.01.011.
Nuclear Energy Agency (NEA), 2019, The Costs of Decarbonisation: System Costs with High Shares of Nuclear and Renewables, https://www.oecd-nea.org/jcms/pl_15000. Accessed Oct. 2022.
Nuclear Energy Institute (NEI), 2021, Nuclear Costs in Context, https://www.nei.org/CorporateSite/media/filefolder/resources/reports-and-briefs/Nuclear-Costs-in-Context-2021.pdf. Accessed Oct. 2022.
Financing
The challenges of financing nuclear power plants are daunting. The entire nuclear ecosystem took note when Westinghouse, a century old US group, was quite literally bankrupted by sinking over US$9 billion into a failed nuclear power project. This project was affected by the sadly common delays and cost overruns in nuclear energy (as with most megaprojects from defense to infrastructure to be fair). And contemporary nuclear energy stakeholders are all well aware of the large challenges posed by a nuclear power plant’s financing more specifically. As such, one might even say that financial innovation will be every bit as important in determining the future of the nuclear industry, as technology and public opinion will.
Indeed, as we’ve seen already, by far the largest part of a nuclear power plant’s capital cost is its financing. And more specifically the interest part of it:
Source: UK National Audit Office as restated in Threadneedle, 2022.
We see that we need to talk about interest. What drives the absolute level of interest in a nuclear project?
1. National interest rates. By definition this is the first building block of any commercial interest rate. And in the current environment of rising interest rates, this is making the financing of nuclear power plants in the coming years more difficult, not easier.
2. The capital amount needed. Interest is of course expressed as a percentage of the total amount borrowed. So the lower the latter, the lower the interest charge for a project as well. It is here that most companies are focusing their efforts when it comes to reducing the cost of nuclear power plants. Through innovative designs and manufacturing methods, they aim to significantly reduce the capital cost of constructing a nuclear power plant, which will also reduce the hefty interest charge on these projects.
3. Interest is also a factor of perceived risk. One would think that with their scale and private and public multi actor backing, nuclear power plant projects are seen as low risk by financiers, but this is certainly not the case in most of the developed world. The main components of this risk are typically:
- revenue uncertainty for the project. Nuclear projects can take 5–15 years to build. In deregulated electricity markets, it is impossible to know, let alone guarantee, what the price of electricity will be in 15 years when having invested US$5–10 billion in your nuclear power plant, you can finally start generating revenue and recouping your costs
- the risk of cost overruns, which plagues the economics of nuclear power projects. Nuclear power projects can cost >3 times more than expected, based on recent builds in the developed world (Rogner et al. 2021)
- the risk of delays. Here too, nuclear builds have regularly taken up to 3 times longer than forecasted from first pour of concrete, to connection to the electricity grid. If you assume that you will start generating revenue in 5 years but you end up only “launching” in 12–15 years, once again you can imagine what this will do to your project’s economic foundation. For example the Generation III reactors in France and Finland — while first of a kind (FOAK) projects it is true, nonetheless took 3.5 times longer than expected to complete (Rogner et al. 2021)
The nuclear industry certainly doesn’t have any influence on the first point here, the national interest rates. Which leaves nuclear stakeholders focused on reducing the second and third points.
Regarding the third point, which is the focus here, the financing of nuclear power plants today goes from fully public, a model often used in China for example, to fully private, typically seen in developed markets. This spectrum from public to private includes:
· public private partnerships (PPP).
· build-operate-transfer (BOT),
· build-own-operate (BOO),
· corporate balance sheet financing,
· vendor equity and
· export credit agency financing
Note: Widely used in infrastructure and energy infrastructure, project finance has not been successfully used in nuclear power due to the size and duration of the financing required (WNA, 2020 at https://world-nuclear.org/information-library/economic-aspects/financing-nuclear-energy.aspx)
However in spite of these existing mechanisms, due to the sheer scale and length of the projects involved, private industry in many developed countries has struggled in recent years to build nuclear power plants on its own. Indeed, the “natural” party to mandate, build, and operate a nuclear power plant, is a large, regional or national, utility company. However, such projects have proven too big even for the largest such entities. Meaning that at this point in time, in particular as in developed countries new nuclear builds are driven by FOAK Generation III reactors, we typically find a government somewhere in the financing value chain.
In addition to lowering the cost of construction, the nuclear energy industry needs financial innovation aimed at derisking these large capital projects, directly or indirectly lowering their perceived risk and financing costs, rendering them viable for market participants to undertake.
A few such examples from around the world:
The most obvious way to reduce a project’s risk is for a government to simply provide the loans/financing necessary to a project. And indeed in terms of new builds in developing markets, the financing mechanism predominantly involves funding help from the government whose companies, public or private, are building the nuclear power plant (the exporting country). Russia and China for example regularly provide up to 90% of the cost of building nuclear power plants abroad by their companies. The French, South Korean, and US governments have similarly provided up to US$10 billion in financing for foreign nuclear projects built by their companies in foreign countries (developing countries but not only — France and Sweden, directly and indirectly, both participated to the financing of the EPR III in Finland for example)
A less involved way to reduce risk is for governments not to provide the loans itself, but to partially or fully guarantee a project’s loans. In the US, Congress approved US$18.5 billion in loan guarantees for the nuclear industry (of the 122 billion requested by the industry, and the 54 billion recommended by President Obama), which were used to implicitly back the financing of the Generation III Vogtle power plant reactors, 3&4.
Another way of reducing risk is to provide guarantees regarding future revenue streams. This can take the form of a typical purchase power agreement (PPA). This is for example what was used to build the Akkuyu power plant in Turkey, whereby a PPA with a Turkish power wholesaler guaranteeing an average price over 15 years for the majority of the plant’s output.
Another means of reducing risk is to borrow from the hedging instruments used in the commodities world for a century or more. In 2014, the UK introduced the contract for difference (CfD), widely used in commodities and finance, to the energy market in order to support the development of clean energy. This is effectively a hedging mechanism, between the nuclear owner/operator, and consumers, whereby consumers guarantee the operator a minimum price (they compensate the operator for the difference if electricity costs drop below this level — a strike price that includes costs and a margin). And consumers retain the difference if the price of electricity is above this same level (the operator reimburses consumers for this difference/excess). This form of financing has been used at the new nuclear reactor at Hinkley Point C for example, with a strike price of GBP 92.50 / MWh as a reference. But the CfD model has been seen as only partially successful in reducing costs at Hinkley Point, with the cost of capital still representing two thirds of the overall cost. As such, the UK government is now introducing another innovative model (innovative for the nuclear industry), the regulated asset base model (RAB), for future nuclear power plants such as the one planned at Sizewell (WNA, 2020).
Another pioneering financing model that deserves note here is the Finish Mankala model, also tried in France as “Exceltium.” In what has been termed cooperative finance, a group of companies, typically heavy industry ones/large electricity users, get together and jointly finance the construction and operation of a new nuclear power plant. Their joint ownership of the project gives them the right (and obligation) to purchase its electricity in the future. This fully private model successfully financed all of Finland’s nuclear industry since the 1970s. Under Exceltium in France, the model aimed similarly for industries, banks and the national utility EDF to cooperate on bearing the costs of new nuclear builds, with investors entitled to buy (and resell if they wished) energy from the projects they financed, with an expectation to recoup their investments over a 24 year period.
The latest development in financing the nuclear industry emerged from France barely a few months ago. Here, with the blessing of the European Union which declared nuclear energy a sustainable source of energy, EDF, the French national utility, issued the first green bonds backed by nuclear energy (https://www.edf.fr/en/the-edf-group/dedicated-sections/investors-shareholders/bonds/green-bonds). Given how large sustainable finance and investing has become, the successful inclusion of nuclear energy into this blossoming area of finance has the potential to become a large source of market financing for the nuclear industry.
For a very good and comprehensive overview of the financing of the nuclear industry check out the very good WNA(2020) report on this topic at:
https://world-nuclear.org/information-library/economic-aspects/financing-nuclear-energy.aspx.
Columbia Threadneedle, 2022, Taking a New Look at Nuclear Energy, https://www.columbiathreadneedleus.com/institutional/insights/latest-insights/taking-a-new-look-at-nuclear-energy/details?id=963228c6-6499-458c-93b5-fe6ffebe00a4. Accessed Oct. 2022.
Rogner, H-Holger, et al. “Keeping the Nuclear Energy Option Open.” SSRN Electronic Journal, 2021, https://doi.org/10.2139/ssrn.3778835.
Environment and Nuclear Energy:
Nuclear energy appears as a backbone of the effort to decarbonize our energy systems. Over the lifetime of our nuclear power park globally, nuclear energy is estimated to have avoided a total of 76 Gt CO2 emissions, second only to hydropower in terms of avoided emissions (Gordon 2020, based on IAEA figures). In the USA alone, nuclear energy generates half of all carbon free electricity, and since 1995 it has avoided over 16 billion metric tons of emissions (NEI, 2021).
When we include not just direct emissions but also indirect, Scope 3 emissions, nuclear power has the lowest carbon footprint of any source of energy available today: lower than hydropower, solar, and wind:
Source of data: UN Economic Commission for Europe, 2021 “Life Cycle Assessment of Electricity Generation Options.”
Graph from Quattry et al., 2022
Supporting this point, the IPCC’s 5th Assessment Report (IPCC, 2014) similarly concluded that nuclear energy has the lowest carbon footprint of any source of energy, along with wind turbines (12gCO2/kWh). Hydropower and solar were both significantly higher at 24g and 27–28g CO2/kWh. Coal without carbon capture (CCS) was at 820g, and with CCS at 200g, still almost 20 times higher than nuclear energy.
Furthermore one can also point out to land usage as land itself is an expensive and constrained resource in many countries, and becoming increasingly so. Studies concludes that nuclear power requires 500 times less land than solar power, and 415 times less than wind (Environmental Progress; McCombie and Jefferson, 2016).
Water usage is an area in which energy comes second after only agriculture (however bear in mind that in energy water is typically used for cooling and almost all of it is immediately returned back to its source i.e. a river). On a lifecycle basis, nuclear power plants consume around 2000–3000 liters of water per MWh, similar to coal plants. Solar power requires 6 times less (except for concentrated solar power where usage is higher at 4000 liters per MWh), and for wind power water consumption is immaterial (McCombie and Jefferson 2016).
A last point to take into account here are natural resources. It is often pointed out that a nuclear power plant is, in the end, made of cement, iron, copper, aluminum and silicon, as well as uranium for fuel — all widely available and widely used except for the last one. Solar and wind however depend on neodymium, dysprosium, terbium, europium, and yttrium, all of which are predicted to be in short supply in the near term (IEA, 2021).
Gordon, Jennifer T. Atlantic Council, 2020, International Co-Financing of Nuclear Reactors Between the United States and Its Allies.
IEA (2021). The Role of Critical Minerals in Clean Energy Transitions World Energy Outlook Special
Report, International Energy Agency (IEA), Paris, France.
IPCC, 2014: Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Core Writing Team, R.K. Pachauri and L.A. Meyer (eds.)], https://www.ipcc.ch/site/assets/uploads/2018/05/SYR_AR5_FINAL_full_wcover.pdf
“Key Energy Topics.” Environmental Progress, https://environmentalprogress.org/energy1.
McCombie C and Jefferson M (2016). Renewable and nuclear electricity: Comparison of environmental impacts. Energy Policy 96: 758–769. https://doi.org/10.1016/j.enpol.2016.03.022
Nuclear Energy Institute, 2021, Nuclear Costs in Context, https://www.nei.org/resources/reports-briefs/nuclear-costs-in-context. Accessed Oct. 2022.
What nuclear countries are betting on — assessments and reports:
Nuclear research centers in countries with mature nuclear industries (US, France, UK, Japan…) are by far the largest repository of knowledge, experience and talent in the nuclear field today. It is therefore a good start to review what consensus emerges from these centers regarding the technological future of the nuclear industry.
At a high level, the previously mentioned intergovernmental Generation IV International Forum, selected six designs that it found to be most promising:
1. the Gas-cooled Fast Reactor (GFR),
2. Lead-cooled Fast Reactor (LFR),
3. Sodium-cooled Fast Reactor (SFR),
4. Molten Salt Reactor (MSR),
5. Supercritical Water-cooled Reactor (SCWR, the only light water reactor in this selection, knowing that all roughly existing commercial reactors around the world are LWRs) and
6. Very High Temperature gas cooled reactor (thermal, to differentiate it from the GFR above)
Source: https://www.gen-4.org/gif/jcms/c_40465/generation-iv-systems
In 2015, the French Institut de Radioprotection et de Surete Nucleaire (IRSN) conducted a review of all the Generation IV nuclear systems selected by the GIF above, and concluded that the SFR, or sodium cooled fast reactor, is in their opinion the only Generation IV nuclear reactor mature enough to see a prototype built in the first half of this century.
Source: https://www.irsn.fr/en/newsroom/news/documents/irsn_report-geniv_04-2015.pdf
The UK Nuclear Innovation and Research Office (NIRO), also conducted its own technical assessment of these six designs, and the high temperature gas cooled reactor (including the very high temperature gas cooled reactor), and the sodium cooled fast reactor, are the only technologies that received the highest grade of 7 for maturity, indicating partly that they could realistically be deployed in the first half of this century.
For interest, these were the technological readiness levels as assessed by this UK report:
1. the Gas-cooled Fast Reactor (GFR): 2
2. Lead- cooled Fast Reactor (LFR): 4
3. Sodium-cooled Fast Reactor (SFR): 7
4. Molten Salt Reactor (MSR): 3 (fast) /4 (thermal)
5. Supercritical Water-cooled Reactor: 2
6. Very High Temperature gas cooled reactor (thermal to differentiate it from the GFR above): 7 for the High Temperature Gas Reactor and 5 for the Very High Temperature Gas Reactor
The US Department of Energy seems to concur with both France and the UK at this level. The DoE retained the same six promising technologies as the GIF. However, in a 2017 assessment, it compared the maturity of these different reactor technologies. They concluded that sodium cooled fast reactors and high temperature gas cooled reactors were the most mature technologies. They further added that building a demonstration version of either one would require USD 4 billion and about 13–15 years, of which half (7 years) would go towards construction alone. They did find these two technologies to be deployable post 2030. They found the other technologies to be significantly less mature, meaning they each require 2–4 extra billion just in R&D versus the sodium cooled fast reactor and the high temperature gas cooled reactor, before the same 4 billion being needed to build a demonstration project. And timeline for these other technologies regarding a demonstration reactor was estimated to be significantly longer, 20+ years, and commercial deployment was not seen in the first half of this century as a result (Petti et al. 2017 quoted in Lyman, 2021).
Accordingly, in 2020 the DoE decided to back TerraPower/GE Hitachi’s Natrium, which is a sodium cooled fast reactor, and X-Energy design of a high temperature gas cooled reactor, for its Advanced Reactor Demonstration Program. Here are the main characteristics of the two technologies being backed by the DoE:
In addition, it also selected five designs as part of its risk reduction focus, involving further research on each but without planning any prototypes for them:
The Hermes reactor from Kairos Power (molten fluoride salt/pebble bed fuel technology)
eVinci from Westinghouse, a micro reactor
BANR from BWXT, a micro reactor
SMR-160 from Holtec, a light water small modular reactor
Unnamed prototype from Southern Company (operator of the Vogtle nuclear power plant in Georgia), a molten chloride fast reactor concept
Source: https://nuclearinnovationalliance.org/advanced-nuclear-reactor-technology-primer
What nuclear countries are betting on — global demonstration nuclear park:
Just as interesting as these national assessments, is what countries that are at the edge of civilian nuclear energy, are currently developing. There are over 130 advanced nuclear reactors in various stages of development around the world at this stage (Maria Cristina Odasso, head of business analysis at LIFTT quoted in Billing, 2022). And there are over 30 commercial scale, demonstration reactors around the world (Johnson, 2018). As Ryan Fitzpatrick points out, “In terms of the number of projects, the number of people working on it, and the amount of private financing, there isn’t anything to compare it to unless you go back to the 1960s.” (quoted in Johnson, 2018).
Here is an overview of demonstration projects around the world, and within the US:
Source: https://www.advancednuclearenergy.org/product/advanced-reactors-turning-the-corner
Source: Third Way at https://public.tableau.com/app/profile/third.way/viz/AdvancedNuclearIndustry_TheNextGeneration/Dashboard1
At this level, China has been operating a small scale gas cooled reactor, the HTR-10, for many years — it is the basis of its first two high temperature gas reactors (the only ones operating commercially in the world), which China now operates at its Shidao Bay Nuclear Power Plant in Shandong province. China has also been catching up fast in the field of fast neutron reactors as well. An experimental 65MW reactor, the Chinese Experimental Fast Reactor (CEFR) has been in operation since 2010 near Beijing (and is connected to the grid). And in 2017, China Nuclear National Corporation began the construction of the CFR-600, a 600MW, demonstration unit, sodium cooled fast reactor (operating on MOX fuel) at Xiapu. Construction of a commercial version, the CFR-1000, could start in 2028, with an estimated connection to the grid by 2034.
Source: https://world-nuclear-news.org/Articles/TVEL-to-supply-fuel-for-Chinas-fast-neutron-react
In France, the CEA, EDF, Naval Group and TechnicAtome are jointly developing the Nuward, a small modular reactor based on pressurized water technology (inspired by existing LWR technology effectively; more at: https://www.edf.fr/en/the-edf-group/producing-a-climate-friendly-energy/nuclear-energy/shaping-the-future-of-nuclear/the-nuwardtm-smr-solution/the-solution).
However France ended in 2019 its plans to build an advanced sodium cooled, fast reactor, for demonstration purposes, with the CEA stating that it did not expect Generation IV nuclear reactors to be built in the first half of this century, in order to justify the cancellation after decades and hundreds of millions of euros spent on research for this project.
Source: https://www.reuters.com/article/us-france-nuclearpower-astrid-idUSKCN1VK0MC
In the US, the DoE has been developing the AFR-100, a small modular (SMR), sodium cooled fast reactor (SFR). The DoE is also building its Versatile Test Reactor, a 300 MW sodium cooled fast reactor (which shares core features with TerraPower’s Natrium) in order to support the private sector with R&D on advanced reactor designs. Construction is expected to be finished between 2025 and 2031.
Of all countries, Russia is the clear leader in experience with fast neutron reactors. They have operated a fast sodium cooled unit, the BN-600, for a very long time and more recently introduced the BN-800, a newer and higher capacity version running on MOX (mixed oxide fuel). They are also developing the BN-1220, which will be the commercial installment of this technology, with a stated goal of installing 11GW of nuclear power generation featuring this technology by 2030. They are further developing small modular reactors with a fast neutron spectrum (and lead cooled in this case), the BREST 300 from NIKIET (Dollezhal Scientific Research and Design Institute of Energy Technologies).
And small reactors have attracted a flurry of research and development around the world. LIFTT has identified over 30 SMR designs under development (Maria Cristina Odasso, head of business analysis at LIFTT quoted in Billing, 2022 ). A few noteworthy designs at this level:
The World Nuclear Association presents an exhaustive overview of both Small Modular Reactors (in operation, under construction, and in advanced development with expected construction) (WNA, 2022):
Operational:
Under construction:
And lists a total of 17 other designs that it deems well advanced and ready for near term demonstration.
The WNA also provides an exhaustive overview of advanced reactors (WNA, 2022) although they include Generation III as “Advanced” while in many places including here “Advanced” is used for Generation IV reactors.
When discussing advanced reactors, a relevant point made by Lyman (2021) is that most advanced features exhibited by Generation IV designs, including passive safety features, modular construction, advanced fuels — could be adapted to existing LWR designs. And that some of these were indeed incorporated in existing LWR designs such as the AP1000 or the NuScale. Nuclear companies will regularly have to decide between improving existing reactor designs and well proven and tested technologies, or revolutionizing reactor designs entirely. This touches on the terminology challenges we saw at the beginning — a LWR can look like an advanced Generation IV reactor in theory, and we saw that one of the six designs retained by the GIF was the SCWR, which is based on existing LWR technology.
Billing, Mimi. “Why Nuclear Should Be the next Boom Market, According to Investors.” Sifted, 25 Jan. 2022, https://sifted.eu/articles/next-gen-nuclear-europe/.
Johnson, Nathanael. “Next-Gen Nuclear Is Coming, If We Want It.” Grist, 7 Apr. 2021, https://grist.org/article/next-gen-nuclear-is-coming-if-we-want-it/.
“Small Nuclear Power Reactors.” Small Nuclear Power Reactors — World Nuclear Association, WNA, 2022, https://www.world-nuclear.org/information-library/nuclear-fuel-cycle/nuclear-power-reactors/small-nuclear-power-reactors.aspx.
“Advanced Nuclear Power Reactors.” Advanced Nuclear Power Reactors | Generation III+ Nuclear Reactors — World Nuclear Association, WNA, 2022, https://world-nuclear.org/information-library/nuclear-fuel-cycle/nuclear-power-reactors/advanced-nuclear-power-reactors.aspx.
What startups and their private investors are betting on:
(cont.)
Top nuclear academic institutions:
United States (US News ranking, nuclear engineering programs):
Source: https://www.usnews.com/best-graduate-schools/top-engineering-schools/nuclear-engineering-rankings
World (EduRank, nuclear engineering):
Source: https://edurank.org/engineering/nuclear/
Europe (EduRank, nuclear engineering):
Source: https://edurank.org/engineering/nuclear/
World (Scimago Institutions Ranking, Energy Sciences):
And some of the main nuclear research centers around the world:
*also an International Center Based on Research Reactors (ICERRs). These centers aim to facilitate international collaboration and research by allowing faster and easier access to research reactors and resources to parties that would not necessarily be able to do so otherwise.
Some of the main newsletters in the nuclear energy space:
