Advanced Reactors and International Security

The challenge of climate change has led to efforts to reduce carbon emissions, pushing countries towards a decarbonization of their economy. Reducing the use of fossil fuels in the sectors of transport and heating can be achieved by switching to electricity-based alternatives, e.g. electric vehicles or heat pumps for residential heating. At the same time, the growth of data center services, including developments in the AI sector, has increased electricity demand. To address this increase in demand, there has been a renewed interest in nuclear power due to its low carbon footprint and the ability to provide baseload power plants complementing renewables. However, large conventional nuclear power plants, with reactors producing several gigawatts of electricity, have proven to be economically challenging, with high upfront construction costs and complex construction that can sometimes last decades. The nuclear industry plans to overcome these hurdles through new implementation strategies. Two key concepts in this context are „small modular reactors“ (SMRs) delivering up to 300 megawatts of electricity, as well as „novel advanced reactors“ (NARs) that further differ from existing power plants by using new types of fuel or coolant.¹

Technical Background of Small Reactors & Novel Reactors

To overcome economic hurdles such as high construction costs and extended construction times, SMR manufacturers plan to factory-produce small reactor modules in a modular fashion, to be delivered to an intended power plant location, instead of the current model of building large-scale nuclear power plants at long-term construction sites. By building these modules in a factory, the modules are expected to be lower in cost by re-using machine tooling and production staff across reactor modules. By using standardized and simplified designs, regulatory licensing is also expected to be accelerated by harmonizing certification. These small reactor modules are intended to be transported to a site, where several modules are combined to form a cluster of reactors. These clusters can either consist of many reactor modules (6-12) to produce power comparable to a large conventional nuclear power plant or of a small number of modules (4-6) for a more decentralized grid. This scalability also allows for potential co-location with industrial sites or data centers. Design-wise, the majority of SMRs are scaled down versions of existing commercial reactor designs, including pressurized water reactors (PWRs) as most commonly used in conventional nuclear power plants, often with a design focus on passive safety and tighter integration of components – such as the steam generator – into the reactor module itself.

However, a major technical issue with smaller reactors is a lower fuel economy. Large nuclear power plants are designed to maximize efficiency and fuel utilization, which can be understood as a relationship between a reactor’s volume and surface area. To sustain a nuclear chain reaction, the neutrons liberated during nuclear fission must remain in the reactor core to continue the chain reaction by initiating more fissions. Shrinking a reactor decreases the ratio of volume to surface, allowing more neutrons to make their way to the surface and “leak” from the reactor. To compensate, many SMR designs intend to increase the degree of fuel enrichment, i.e. increasing the amount of easily fissionable uranium isotopes (235U) in the fuel from the usual low enriched uranium (LEU), with 3-5% 235U, to high-assay low enriched uranium (HALEU), with up to 20% of 235U.²

NARs aim to enhance fuel efficiency with new types of fuel or coolants that optimize the neutron economy of small reactors, such as mixed oxide (MOX) or uranium nitride (UN) fuels. Many of these new fuel types also aim for higher enrichment by utilizing HALEU, on which this chapter will focus. New form factors for fuel are also being developed, such as ceramic fuel particles embedded in a graphite matrix, either in the shape of tennis ball-sized pebbles or “prismatic” fuel blocks. The advantage of these new fuel forms is the compatibility with alternative coolants, such as molten salts or inert gases. Water is used as coolant in most large-scale power plants, but since it is a very good neutron absorber, PWR-like designs lose neutrons within the water. Gases, however, are low in density, while the heavier elements making up salts absorb fewer neutrons. These coolants also have the advantage of allowing higher operational temperatures; typically, the critical point of water limits the coolant temperature. The alternative coolants, however, are either already gaseous or have a much higher critical point than water, allowing high temperature operation (>400 °C). Higher temperatures increase the thermodynamic efficiency of a power plant, producing more electricity per unit of heat generated and are beneficial for efficient use of the output heat, e.g. for district heating, desalination or industrial processes.

Finally, some NAR designs consider sealing the fuel inside the reactor vessel for the entirety of operation, as the better fuel efficiency from higher enrichment and neutron economy optimization, combined with a lower electrical output, can allow long-term use (10+ years) without refueling. NARs are technically rooted in previous reactor research, commonly summarized as “Generation IV” nuclear power plants, but as none of these Generation IV designs have been deployed commercially, NARs still require substantial research and development effort.³

Implications of HALEU as Nuclear Fuel

The key difference between LEU and weapons-capable highly enriched uranium (HEU) is the ²³⁵U content in uranium, as only ²³⁵U is fissile, i.e. capable of sustaining the rapidly multiplying chain reaction in a nuclear weapon. Natural uranium consists mostly of ²³⁸U and less than 1% of ²³⁵U. In enrichment facilities, the uranium isotopes are separated to create enriched uranium with a higher fraction of ²³⁵U. Whether producing LEU/HALEU or HEU, the underlying separation techniques are the same, but the separative work required for reaching high enrichment is not linear: producing HEU from HALEU with 20% ²³⁵U requires only ca. 40% of the separative work required to produce HEU from LEU with 5% ²³⁵U as feed material,⁴ illustrated in Figure 1. This raises the overall proliferation risk, as HALEU is more attractive to divert for weapons use and requires less time and resources to convert to military use (“breakout time”). Additionally, past research suggests that HALEU with 20% ²³⁵U can potentially be used to construct a weapon, however, such a device would be significantly larger than current warheads.⁵

Another consideration is the availability of HALEU. At the moment, Russia is the only commercial producer of HALEU, making any HALEU-based reactor reliant on Russian exports. Currently, the United Kingdom is investing heavily into HALEU production facilities,⁷ and the U.S. is pursuing a HALEU availability program.⁸ However, HALEU is not expected to be commercially available from these new facilities for several years. Were these programs to suffer from delays, the supply chain for SMRs would be dependent on a single state. The risk of dependency is especially problematic due to the potential political leverage of a single provider, an undesirable outcome given the adversarial position of Russia in the current geopolitical climate. Conversion of existing facilities to HALEU production is not seen as an alternative, as the assumption of 5% maximum enrichment has shaped facility design and safety procedures and a conversion would therefore require extensive operational reassessment for higher enrichments.⁹

A final consideration regarding HALEU is that the push for higher enriched commercial fuel might place pressure on the non-proliferation efforts. For example, limits were placed on Iran as part of the JCPOA,¹⁰ restricting Iran’s uranium enrichment to 3.67%. If HALEU becomes the new “commercial standard”, placing enrichment limits below HALEU levels in any international agreements will likely become less viable.

Security and Safeguards Challenges of Small Facilities

The plans for SMR deployment foresee many smaller nuclear sites, potentially co-located with a facility requiring electricity (such as a data center) or heat (chemical factory, desalination). The change to smaller sites has direct consequences for the physical security of sites to prevent unauthorized access to nuclear material and the safeguards regime for detecting clandestine diversion of fissile material. Approaches to both problems have traditionally relied on the current model of nuclear facilities: a relatively small number of reactors and fuel storage located in dedicated locations with multilayered security barriers surrounding the facility.¹¹ The potential increase in the total number of facilities as well as a smaller security perimeter means adequate surveillance and verification inspections require more personnel and effort to maintain current response and detection times.

Another complication arising from the increase in nuclear sites is safety and security during armed conflicts. During the Russian invasion of Ukraine, the occupation of the Zaporizhzhia Nuclear Power Plant raised major international concerns.¹² With an increased number of nuclear sites, the likelihood of nuclear sites being similarly involved in armed conflicts also increases, leading to a higher risk of accidental or deliberate radiological release.

The co-location with other facilities, as well as the use in remote locations – such as mining communities – further increases the logistical demands on a safeguards inspectorate. This is especially critical because the International Atomic Energy Agency (IAEA) that maintains the safeguards regime is subject to strict budgetary constraints. One strategy to mitigate this issue is the increased use of remote monitoring systems for safeguards. However, because this requires remote data transmission to the inspectorate headquarters, a potential side effect is a larger attack surface for cyberattacks. If reactors are intended to be internationally transported between manufacturing and operating state without unsealing, then mobile reactor concepts or designs with fuel sealed inside the reactor vessel will likely require new inspection protocols, raising the question of responsibility and liability for material accountancy or discrepancies between inspections.

Finally, the physically smaller size of the proposed SMRs/NARs means a higher density of reactor components, making material sampling or quantitative measurements more challenging. Similarly, the new materials proposed by NARs (fuel pebbles, liquid fuel, molten salt coolants) also mean current verification techniques might not be applicable to these future designs.¹³

Better Security & Safeguards – by Design

The challenges to security and safeguards posed by these new developments can be addressed on several levels: Firstly, additional consideration should be given to SMRs and NARs that can achieve their goals without HALEU or other fuel types with additional proliferation concerns, requiring no – or only – a moderate increase of enrichment. Secondly, there must be research and development to develop new verification techniques, – or to adapt existing techniques – for new reactor types, to prevent any gaps in capability. This includes the development of techniques to verify reactor operation and fuel composition more quickly, reliably, and precisely, to reduce the logistical burden imposed by the increased number of facilities. Thirdly, because most of these new reactor designs are still in the research and development phase, there is the chance to integrate new technology and regulatory requirements into the design process. However, this requires timely engagement with the reactor designers and supporting existing initiatives, e.g. “safeguards-by-design”.¹⁴ Ideally, these aspects need to be addressed by all stakeholders during the entire process (investment, research and development, site selection, site licensing), which should ideally see the future users and operators (and not only reactor developers) included early in the process.