Robert Eccles
Robert G Eccles is at the Saïd Business School, University of Oxford.
It’s been a big year for nuclear energy in the US. The Department of Energy has allocated a large amount of capital to nuclear energy research and has committed $900 million to advance Gen III+ (more on them below) small modular reactors (SMRs). The Inflation Reduction Act’s inclusion of nuclear energy has opened opportunities for tax credits for investors in nuclear projects. Southern Company’s Vogtle plant’s second new reactor started sending power to the grid in April.
Most recently, on September 20, Microsoft and Constellation announced that they will reopen a reactor at Constellation’s Three Mile Island nuclear energy center in Pennsylvania to power Microsoft data centers. Microsoft agreed to pay $16 billion to restart the Unit 1 reactor which has a capacity of 835 megawatts. It was shut down in 2019 under financial pressure from growing competition with cheap natural gas. (The Unit 2 reactor was destroyed in 1979 accident and is undergoing decommissioning; however, the US Nuclear Regulatory Commission (USNRC) noted there were no deaths or accidents, or discernible health effects from some small radioactive releases.) Microsoft has agreed to buy up to 100 per cent of the electricity produced by Unit 1. This is part of the tech giant’s efforts to secure enough reliable, low-carbon electricity to supply its energy-thirsty data centers powering the boom in artificial intelligence.
An Overview of Nuclear Energy
For many years, nuclear energy existed in the shadows of failed projects and safety concerns. While solar and wind flourished and fossil fuels remained a standby, the nuclear power industry has made huge strides in increasing safety, commercial feasibility, and attractiveness of nuclear as a carbon-free, reliable, baseload power source. It’s no surprise that Microsoft is now making headlines as it secures nuclear power to fuel its growing AI demand.
The 2000+ page Sixth Assessment Report (AR6) on the mitigation of climate change by the Intergovernmental Panel on Climate Change notes that nuclear energy capacity must nearly double by 2050 in order to keep the global temperature rise below 1.5°C. Nuclear power provides carbon-free baseload power without the intermittency inherent in wind and solar power. This report was published in 2022 and since then we have witnessed the growth of AI and its insatiable energy needs. AI is putting at risk all the net-zero commitments made by the big tech companies. This makes nuclear energy even more important.
Yet, as pointed out by the Nuclear Innovation Alliance, the 42-page AR6 Synthesis Report’s Summary for Policymakers only mention nuclear once and “that single mention is squeezed within a densely populated table that in no way acknowledges its importance for greenhouse gas reduction.” The reason for this is that some of the countries that determine what is in the summary decided to downplay the important role of nuclear energy, an omission that was opposed by the US.
This ambivalence about nuclear energy is not surprising, but attitudes are changing. A 2024 survey by Pew Research Center found that 56 per cent of Americans favor more nuclear power plants to generate electricity (compared to 78 per cent for solar and 72 per cent for wind). Judi Greenwald, the Executive Director of the Nuclear Innovation Alliance, has recently written about how environmentalists have changed from an anti-nuclear to a pro-nuclear position over the past 40 years as they have come to better understand its benefits and realize the risks are extremely small and easily mitigated. In this short video Nick Loris of the center-right climate think take C3 Solutions explains that the perceived risks are much less than the actual ones.
Last year ClearPath, another center-right climate think tank, published a report based on a survey it did with a group of center-right and center-left organizations. The survey covered 13,500 respondents in eight countries: France, Germany, Japan, Poland, Sweden, South Korea, the UK, and the US. On average there were five supporters for every opponent, with the range of support across countries being 65-92 per cent. Respondents saw the benefits and were not put off by issues of cost, safety, and waste. Environmentalists support nuclear, support exists across the political spectrum, and the young are particularly receptive,
Support could be even greater with a better understanding of nuclear energy. In particular, an understanding of small modular reactors (SMRs) since this is the nuclear technology that can best deliver on the need for nuclear energy. As defined by the International Atomic Energy Industry (IAEI), SMRs are advanced nuclear reactors that have a power capacity of up to 300 MW(e) per unit, which is about one-third of the generating capacity of traditional nuclear power reactors.” SMRs are faster (3-4 years vs 10) and cheaper (about half the cost per megawatt hour) to build due to a factory-controlled manufacturing environment, have a much smaller emergency planning zone (40 acres vs a 10-mile radius), and are safer than traditional nuclear (which, properly managed, is very safe).
For some basic background on SMRs see this video discussion of SMRs by US Department of Energy Principal Deputy Assistant Secretary Ed McGinnis, tutorials by ClearPath, and a video by the IAEA on the safety and security of SMRs. It is also worth noting that the European Commission supports SMRs. “The Commission recognises the potential contribution of Small Modular Reactors to achieving the energy and climate objectives of the EU Green Deal, as reflected in its recommendation for the 2040 emission reduction targets.”
According to the World Nuclear Association today there are around 440 commercial nuclear reactors supplying nine per cent of the world’s electricity and one-quarter of the world’s low-carbon energy, with 64 reactors under construction globally. These are conventional nuclear designs and not small modular reactors where approximately 70 per cent of their designs are based on pressurized water reactors and the remaining are based on boiling water reactors and heavy water reactors.
These two technologies are also used for SMRs. Here is a brief review of each design.
Pressurized Water Reactors (PWRs)
Nuclear fission produces heat inside the reactor. That heat is transferred to water circulating around the uranium fuel in the first of three separate water systems. The water is heated to extremely high temperatures but doesn’t boil because the water is under pressure. The water within the primary system passes over the reactor core to act as a moderator and coolant but does not flow to the turbine. It is contained in a pressurized piping loop. The hot, pressurized water passes through a series of tubes inside the steam generator.
These tubes are surrounded by another water system called the secondary or steam generating system. The heat, but not the water, from the primary coolant is transferred to the secondary, system which then turns into steam. The primary and secondary systems are closed systems. This means the water flowing through the reactor remains separate and does not mix with water from the other systems. The steam is pumped from the containment building into the turbine building to push the giant blades of the turbine. The turbine is connected to an electrical generator.
After turning the turbines, the steam is cooled by passing it over tubes carrying a third water system called the condenser coolant. As the steam is cooled, it condenses back into water and is returned to the steam generator to be used again and again.
PWR technology brings the advantage of using similar fuel to what is used in the global fleet of conventional commercial reactors of Low Enriched Uranium (LEU) which has less than five per cent enrichment. (More on this below.) Also, these designs are very stable due to their tendency to produce less power as temperatures increase. In these designs, a leak of radioactive nuclides in the core would not transfer any radioactive contaminants to the turbine and the condenser. One of the disadvantages is their lack of commercial deployment.
US SMR companies using PWR technology include:
For more detail on PWRs see this explanation by ScienceDirect.
Boiling Water Reactors (BWR)
Unlike the PWR, inside the boiling water reactor the primary water system absorbs enough heat from the fission process to boil its water. In contrast to the PWR, the BWR uses only two separate water systems as it has no separate steam generator system. The result of this is that the steam is contaminated, explaining why and this design represents only about 15 per cent of the market. This steam and water mixture rises to the top of the reactor and passes through two stages of moisture separation. Water droplets are then removed, and steam is allowed to enter the steam line. The steam is directed to the turbine. The turbine begins to turn within the generator and electricity is produced.
Once the turbines have turned, the remaining steam is cooled in the condenser coolant system. This is a closed water system. Heat from the steam is absorbed by the cool water through heat transference. The water within the two systems does not mix. Once through the condenser system, the water is recycled back into the reactor to begin the process again.
BWR technology is more simplified and requires fewer components than the indirect cycle of its counterpart. BWR plants also operate at a lower nuclear fuel temperature. One of the disadvantages with the BWR design is that the control rods are inserted from below. There are two available hydraulic power sources that can drive the control rods into the core for a BWR under emergency conditions. Most other reactor types use top entry control rods that are held up in the withdrawn position by electromagnets, causing them to fall into the reactor by gravity if power is lost.
US companies using BWR technology include:
For more detail on BWRs see this explanation by ScienceDirect.
Enriching Uranium
In addition to the technology being used, SMRs are classified as Gen III, Gen III+, and Gen IV. Discussing the differences between them requires a basic understanding of uranium enrichment. (More detail can be found on the USNRC’s website.) The nuclear fuel used in a nuclear reactor needs to have a higher concentration of the U235 isotope than that which exists in natural uranium ore. U235 when concentrated (or “enriched”) is fissionable in light-water reactors (the most common reactor design in the USA). During fission, the nucleus of the atom splits apart producing both heat and extra neutrons. Under controlled conditions, these extra neutrons can cause additional nearby atoms to fission and a nuclear reaction can be sustained. The heat energy released by the controlled nuclear reaction within the nuclear reactor can be harnessed to produce electricity. Commercially, the U235 isotope is enriched to three to five per cent (from the natural state of 0.7 per cent) and is then further processed to create nuclear fuel.
At the conversion plant, uranium oxide is converted to the chemical form of uranium hexafluoride (UF6) to be usable in an enrichment facility. UF6 is used for two reasons. First, the element fluorine has only one naturally-occurring isotope which is a benefit during the enrichment process (e.g., while separating U235 from U238 the fluorine does not contribute to the weight difference). Second, UF6 exists as a gas at a suitable operating temperature.
The two primary hazards at enrichment facilities include chemical hazards that could be created from a UF6 release and criticality hazards associated with enriched uranium.
There are several enrichment processes utilized worldwide. They are:
Gen III, Gen III+, and Gen IV
These reactor designs are considered to be Gen III designs. The Gen III+ designs are similar but with improvements to safety and performance. Generation III and III+ generally use Low Enriched Uranium (LEU) fuel which is enriched between three to five per cent. Nature is comprised of the U-235 isotope (at around 0.7%) and the U-238 isotope. The only useful isotope is U-235 and so there is an enrichment process to get to three to five per cent. This type of fuel has also been used in the global conventional commercial plants for more than 50 years providing strong operational data and performance data understood by technology companies and regulators. This type of fuel normally allows for an 18 to 24 month refueling cycle per reactor.
Gen IV reactors are generally advanced or new design types not yet in commercial operation and also have not been fully designed or developed. Instead of using LEU, Gen IV designs use High Assay Low Enriched Uranium (HALEU) which is enriched to five to 20 per cent. Advantages for this include the refueling cycle being extended past 24 months, consuming more fuel to reduce the amount of used fuel at the end of the reactor life, and smaller reactor cores.
One of the main challenges for these designs is that HALEU is not yet widely available commercially. At present only Russia and China have the infrastructure to produce HALEU at scale and Russia is more predominant commercially. In light of the current situation with Russia, this fuel is not available to the U.S. market. The center-left think tank Third Way has pointed out the vulnerability this creates for operators of nuclear facilities in the US, Europe, and elsewhere. Currently, there are measures being implemented to produce HALEU in the US, but this will take significant funding and time. In the meantime, H.R. 1042, “Prohibiting Russian Uranium Imports Act,” became public law on May 13, 2024.
US companies developing the Gen IV technology include:
Licensing a Nuclear Power Plant in the US
Both existing and new technologies require getting a license from the USNRC. There are two codes within the NRC’s regulations that provide the process to build a nuclear power plant. The “operator” is the owner in some cases (but not all) of the plant and the “technology supplier” is the manufacturer.
One process is 10 CFR Part 52. This process is preferable for operators because most of the risk and cost is borne by the technology supplier upfront in the licensing process with the Standard Design Certification (a commission approval, issued under subpart B of this Part 52, of a final standard design for a nuclear power plant), where:
- The bulk of construction activities only start after the design is complete, certified, and the Combined Operating License (US Nuclear Regulatory Commission (NRC) authorizes the licensee to construct and (with specified conditions) operate a nuclear power plant at a specific site, in accordance with established laws and regulations) is issued
- A public hearing before operation requires a prima facie finding that one or more inspections, tests, analyses, and acceptance criteria (ITAAC) have not been met
The second process is 10 CFR Part 50 which is a two-step process and was primarily used before the more streamlined Part 52 process was developed. Most of the regulatory work and cost is borne by the owner/operator, where:
- There is greater risk on the customer due to allowing construction to begin upon issuance of the Construction Permit, where designs are immature and prone to change, potentially increasing construction timeframes and costs
- A second hearing after construction may stall operation
There is yet another process for the Gen IV technologies or technologies that are not yet operating commercially. This process requires the technology supplier to construct a demonstration plant, where this plant will serve as a facility to prove the operation and collection of fuel data. It is required that the demonstration plant operate for at least two fuel cycles in order to have sufficient fuel data to provide to the USNRC for them to begin their technology reviews. Once this has been obtained the USNRC can begin their review of the technology. Once this has been completed (duration for review is undetermined but is typically four years or more), a two-year period is required for public hearings. Then the decision will be made by the USNRC if this demonstration plant can be considered and operated as a commercial plant or does a new commercial plant need to be constructed. The total time period is easily 10 years. The obvious corollary is that it will be at least a decade before any of the Gen IV companies will be supplying nuclear energy in the commercial market.
Current Commercial Capacity
Today there are numerous SMR technologies in development at various stages of completion. Only NuScale has received USNRC approval/certification to manufacture and sell its SMR. This took the company a record 42 months (based on PWR technology) to achieve the approval/certification at a cost of $500 million. On top of the years and cost of developing and testing the technology. In addition, NuScale has formed a joint venture entity and an exclusive global strategic partnership with ENTRA1 Energy to sell and install NuScale SMRs in the US and globally via “ENTRA1 Energy Plants™ + NuScale SMRs-inside,” offering a unique and immediate one-stop-shop solution for SMR commercial deployment and power production.
The day before the Three Mile Island announcement, an article in TechBullion reported that “NuScale Power Corp is set to play a pivotal role in powering Microsoft’s AI ambitions.” This is in the context of a $30 billion fund being led by Microsoft and BlackRock to support the development of AI technologies and the energy needed to power them. The article elaborates that “A collaboration between Microsoft, BlackRock, and NuScale could reshape the landscape of both AI and energy.”
While this is very encouraging, the US will need more SMRs than one company can make. This raises some challenging questions regarding where other companies stand in their development efforts and the process used by the USNRC to grant licenses. It’s important to ensure safety while at the same time expediting the development of SMR capacity in the US.
The nuclear energy industry encompasses a range of options, from the more traditional plants like Vogtle and Three Mile Island to SMR-based solutions that vary in technology and fuel and are at a range of stages of development. It is no surprise that technology companies are at the forefront of adoption. With AI and other advanced data processing requiring huge amounts of energy around the clock, and technology companies making bold decarbonization plans, the current energy mix just won’t work. Nuclear energy is not a silver bullet solution, but can be a significant part of our future domestic energy mix. Large energy users, developers, policymakers, and financiers should be getting up the learning curve and spending time working on solutions to expedite a range of nuclear projects. A thriving and clean domestic energy market requires it.
This article originally appeared at Real Clear Energy and was republished by CFACT.
