Small Modular Reactors: Serial Production Versus Nuclear Physics

For most of the last fifty years, the trajectory of nuclear power has followed a singular logic: bigger is better. Engineers pursued massive gigawatt-scale reactors on the premise that doubling the size of a pressure vessel does not double its cost. This pursuit of economies of scale resulted in the colossal concrete domes that define the current fleet. However, the financial reality has diverged from the engineering theory. These projects have become so complex that they are frequently crippled by decade-long delays and multibillion-pound budget overruns.

The industry is now attempting a fundamental pivot toward Small Modular Reactors, or SMRs. These are fission reactors with a power output typically between 50 and 300 megawatts, designed to be manufactured in sections within a factory and transported to a site for final assembly. The ambition is to replace the bespoke, site-specific nature of traditional nuclear construction with the repeatable precision of an aircraft assembly line. If the economics of scale have failed, the industry hopes that the economics of serial production will succeed.

This transition involves more than just shrinking a large reactor. It requires a total reconfiguration of the safety systems and the underlying physics of the core. In a traditional large reactor, the heat generated by the fuel is so intense that it requires a complex array of pumps, backup generators, and active cooling systems to prevent a meltdown in the event of a shutdown. SMRs aim to simplify this through passive safety, using the natural laws of thermodynamics to cool the reactor without human intervention or external power.

Whether this shift can actually lower the cost of electricity remains an open question. While factory production reduces the risk of construction delays, it introduces new challenges in neutron efficiency. A smaller reactor core inherently loses more neutrons through its surface than a large one, which can affect the longevity and efficiency of the fuel. The success of SMRs depends on whether the gains from manufacturing can outweigh these inherent physical disadvantages.

The departure from gigawatt-scale architecture

The current generation of large-scale reactors, such as the EPR or the AP1000, represents the pinnacle of twentieth-century nuclear design. These plants are designed to provide a massive, steady flow of electricity to the national grid. Because the cost of the site, the security, and the regulatory oversight is high regardless of the reactor's power, building a 1,600-megawatt unit is generally more cost-effective on paper than building two 800-megawatt units.

However, the sheer physical size of these components creates a bottleneck. A reactor pressure vessel for a gigawatt-scale plant can weigh over 500 tonnes and must be forged by one of only a handful of facilities worldwide capable of handling such massive steel ingots. Transporting these components requires specialized heavy-lift infrastructure. Once at the site, the project becomes a massive civil engineering undertaking, involving thousands of workers and decades of coordination. The complexity is so high that a single error in a weld or a delay in a concrete pour can cascade into years of lost time and millions in interest payments.

SMRs represent an intentional retreat from this scale. By limiting the size of the reactor, designers can use smaller, more common industrial components. The pressure vessel of an SMR is small enough to be transported by rail or barge. This allows the majority of the high-precision work to be performed in a controlled factory environment, where quality control is easier to maintain and weather does not interfere with the schedule. The goal is to move from "project-based" nuclear to "product-based" nuclear.

The transition to modularity seeks to replace unique civil engineering feats with the repeatable efficiency of a manufacturing plant.

This modularity also allows for a different financial model. A utility could theoretically install one small module to start generating revenue, then add subsequent modules as demand or capital allows. This reduces the "all-or-nothing" risk that has made private investors wary of large-scale nuclear projects. By decoupling the generation of power from the massive scale of the site, SMRs aim to make nuclear power more flexible and less financially catastrophic.

Neutron economy and the physics of scale

While the financial arguments for SMRs are compelling, the physics of a smaller core presents a significant engineering hurdle known as neutron leakage. Nuclear fission relies on a self-sustaining chain reaction where neutrons released by one splitting atom go on to hit another. To maintain this reaction, a certain density of neutrons must be maintained within the fuel.

In any reactor, some neutrons will inevitably escape the fuel region and be absorbed by the surrounding structural materials or the shielding. The efficiency of a reactor is largely determined by the ratio of the core's volume to its surface area. A large reactor has a high volume-to-surface-area ratio, meaning a smaller percentage of its neutrons reach the edge and escape. As the core is scaled down, the surface area becomes larger relative to the volume, and neutron leakage increases.

To compensate for this leakage, SMR designers have several options, none of which are without cost. They can use fuel that is more highly enriched, meaning it contains a higher concentration of the fissile isotope Uranium-235. While this sustains the chain reaction, it increases the cost of the fuel and may require more stringent security measures for transport and storage. Alternatively, they can use more effective neutron reflectors—materials like beryllium or graphite that bounce neutrons back into the core—which adds complexity and cost to the vessel design.

Another consequence of a smaller core is the fuel cycle length. Large reactors can typically run for 18 to 24 months before they need to be powered down for refueling. Because SMRs have less total fuel, they may need to be refueled more frequently, or they may need to operate at a lower power density to stretch the fuel life. Some designs, such as those using molten salt or lead cooling, attempt to solve this by allowing for continuous refueling or by using a "long-life" core that lasts for a decade but is then replaced entirely. These technologies remain largely in the demonstration phase and have yet to be proven at a commercial scale.

Principles of passive heat removal

The most significant safety advantage of SMRs lies in the physics of heat removal. When a nuclear reactor is turned off, the radioactive decay of fission products continues to generate heat. In a large reactor, this "decay heat" is so substantial that if the cooling pumps fail, the fuel can melt within hours. This is why large plants require multiple, redundant active cooling systems, often powered by diesel generators or massive battery banks.

SMRs are designed to rely on passive safety systems, which use natural phenomena like convection, conduction, and gravity rather than powered pumps. Because the total amount of decay heat is much lower in a small reactor, it is easier to manage. In many SMR designs, the reactor vessel is submerged in a large pool of water. If the plant loses all external power, the water in the pool absorbs the heat from the reactor. As the water heats up, it naturally rises, moves to a heat exchanger, cools, and sinks back down, creating a continuous loop of cooling without any moving parts.

• Natural circulation: The density difference between hot and cold water drives the coolant through the core without the need for mechanical pumps.
• Large heat sinks: The ratio of the reactor’s surface area to its thermal output is high enough that heat can dissipate through the vessel walls into a surrounding tank or the atmosphere.
• Simplified containment: The smaller size allows for a containment structure that can be kept at high pressure or submerged, further reducing the risk of a leak.

This shift to passive safety does not just make the reactor safer; it simplifies the entire plant. By removing the need for safety-grade pumps, complex piping networks, and backup generators, designers can reduce the total number of parts in the plant by a significant margin. This reduction in complexity is a prerequisite for factory assembly. However, the regulatory challenge remains. Existing safety standards were written for large, active systems, and proving to a regulator that a passive system is equally reliable requires extensive testing and simulation.

From civil engineering to factory assembly

The ultimate goal of the SMR movement is to move nuclear power into the realm of advanced manufacturing. In a traditional build, the site is a chaotic environment where thousands of variables—from soil quality to local labor disputes—can impact the project. In a factory, the environment is controlled. Specialized jigs and robotic welding systems can perform tasks with a precision and speed that is impossible in the field.

In this model, the reactor is divided into several modules: the core vessel, the steam generators, the containment structure, and the control systems. Each module is built on a separate line and then transported to the site. This allows for parallel construction; while the site is being excavated and the foundation poured, the reactor itself is already being manufactured hundreds of miles away. The time spent on-site is reduced from a decade to perhaps three or four years.

This approach relies on the concept of the "learning curve." In manufacturing, the cost of a product typically drops by a certain percentage every time the cumulative production volume doubles. This has been seen in the production of solar panels and wind turbines, but it has never been successfully applied to nuclear power, where every plant has been a unique "first-of-a-kind" project. To achieve these savings, the industry must commit to a high degree of standardization. If every customer demands slight changes to the design, the benefits of the assembly line vanish.

There is also the challenge of the supply chain. Transitioning to SMRs requires a new network of suppliers capable of producing high-spec nuclear components at a steady rhythm. Current published figures for SMR orders are modest, and without a large backlog of orders, manufacturers are hesitant to invest in the expensive tooling required for mass production. The industry is currently in a "chicken and egg" situation: costs will only drop once production reaches a certain volume, but that volume will only be reached if the initial costs are low enough to attract buyers. The first few commercial SMRs, currently in various stages of planning and early construction in North America and Europe, will serve as the crucial test for this manufacturing thesis.

The learning curve in serial manufacturing

The economic case for small modular reactors (SMRs) rests on the concept of the learning curve, a phenomenon well-documented in aerospace and ship-building but historically absent in nuclear energy. In conventional industries, the cost of producing a unit typically drops by a predictable percentage every time the cumulative production volume doubles. This is achieved through the refinement of fabrication techniques, the reduction of material waste, and the increasing proficiency of the workforce. Large-scale nuclear projects, by contrast, have historically exhibited a negative learning curve. As projects grew in size and complexity, costs tended to rise rather than fall, largely because each plant was effectively a bespoke civil engineering project subject to local site conditions and shifting safety requirements.

To reverse this trend, SMR developers aim to shift the majority of the construction work from the field to the factory. In a controlled environment, components can be forged, machined, and assembled using automated processes that are impossible to implement on a muddy construction site. For instance, electron beam welding can join thick steel sections in minutes rather than the days required by manual arc welding. If a manufacturer can produce dozens of identical reactor vessels, the fixed costs of the specialised tooling and the development of the robotic protocols are amortised across the entire fleet.

However, the steepness of this learning curve remains speculative in the nuclear context. Studies from the International Atomic Energy Agency and various engineering consultancies suggest that a ten to twenty per cent cost reduction per doubling of volume is plausible, yet this requires a level of throughput that the industry has not seen since the height of the Cold War. For a factory to operate at peak efficiency, it must produce multiple units per year. If the order book is thin, the factory sits idle, and the high overheads of maintaining a nuclear-certified facility quickly erode any gains made through serial production. The success of the modular approach depends less on the physics of the reactor and more on the logistics of the assembly line.

Supply chain constraints for specialised alloys

While the assembly line promises efficiency, it is entirely dependent on a reliable flow of raw materials and intermediate components. SMRs, particularly those designed to operate at higher temperatures or using non-water coolants like molten salts or liquid sodium, require advanced materials that are not currently produced at scale. High-chromium steels, nickel-based superalloys, and specialised zirconium alloys are necessary to withstand the intense neutron flux and corrosive environments inside a compact core.

The global supply chain for these materials is currently brittle. Only a handful of forges worldwide possess the capacity to produce the large, high-purity ingots required for reactor pressure vessels. Even for smaller modular designs, the quality standards are exacting. Every batch of steel must be traceable to its heat of origin, with documented chemical composition and mechanical properties. This level of oversight creates a significant lead time. A delay in the delivery of a single specialised valve or a specific grade of welding wire can halt an entire assembly line, a risk that is magnified when multiple reactors are being built simultaneously.

Furthermore, the shift toward High-Assay Low-Enriched Uranium (HALEU) presents a fuel-cycle bottleneck. Many advanced SMR designs require uranium enriched to between five and twenty per cent to maintain a compact core size and extend the period between refuelling. At present, commercial production of HALEU is extremely limited. Until a robust, international supply chain for this higher-enriched fuel is established, the deployment of next-generation SMRs will remain restricted to experimental or state-subsidised projects. The industrial infrastructure required to support a fleet of SMRs must be built in tandem with the reactors themselves, representing a secondary, multi-billion pound investment that is often overlooked in promotional literature.

The transition from bespoke craftsmanship to industrialised assembly requires a fundamental shift in how the nuclear industry values precision and repeatability.

Regulatory hurdles for standardised designs

The traditional regulatory model is based on site-specific licensing. Regulators assess the geology, hydrology, and population density of a specific location before granting a permit to build a specific reactor design. For SMRs to be economically viable, this process must be streamlined. A manufacturer cannot afford to modify the internal layout of a standardised module to satisfy the idiosyncratic requirements of different national regulators. The ideal is a "design certification" that allows a reactor to be built anywhere within a jurisdiction, or ideally across multiple countries, with minimal site-specific changes.

Progress toward this goal is slow. While the UK Office for Nuclear Regulation and the US Nuclear Regulatory Commission have begun to coordinate on technical reviews, national sovereignty remains a significant barrier. Each regulator has its own standards for redundancy, passive safety features, and cyber-security. A design that is approved in one country may require significant internal redesign to meet the safety culture of another. If a manufacturer has to create three different versions of a "standard" module for three different markets, the economies of scale are lost.

There is also the matter of the "source term" and emergency planning zones. SMR advocates argue that because these reactors have a smaller inventory of radioactive material and enhanced passive safety features, the required buffer zone around the plant should be much smaller than for a gigawatt-scale station. This would allow SMRs to be built closer to industrial hubs or towns to provide district heating. However, regulators are historically cautious. Proving that a small reactor is fundamentally safer to the point of requiring less land is a multi-year evidentiary process. Until these regulatory frameworks are harmonised and modernized, the "modular" nature of the technology remains partially theoretical.

Operational trade-offs in fuel lifecycle

The physics of a smaller core introduces specific operational challenges that differ from those of large reactors. A smaller core has a higher surface-area-to-volume ratio, which leads to greater neutron leakage. To compensate for this, SMRs often use fuel with higher enrichment or different geometric arrangements to maintain the chain reaction. While this allows the reactor to be small, it can result in less efficient fuel utilisation. Over the lifetime of the plant, an SMR may produce more high-level waste per unit of electricity generated than a large-scale pressurized water reactor.

Managing the fuel cycle for a distributed fleet of small reactors also introduces logistical complexity. Instead of a few large sites with deep onsite cooling ponds, a nation might have dozens of smaller sites. This increases the frequency of nuclear material transport and requires a more distributed security apparatus. Some SMR designs propose a "sealed core" model, where the entire reactor vessel is returned to a central facility for refuelling every ten to twenty years. While this reduces the need for onsite fuel handling, it requires a transport infrastructure capable of moving highly radioactive, heavy loads across public rail or road networks.

Decommissioning presents a final trade-off. The modular nature of the reactors should, in theory, make them easier to dismantle. A factory-built module could be disconnected and shipped to a central decommissioning hub, rather than being cut apart on-site. However, this assumes that the transport casks and vessels will be available and affordable decades after the plant is built. If the manufacturer has gone out of business or the design has been superseded, the "modular" advantage may disappear, leaving a legacy of small, dispersed sites that are expensive to remediate.

A smaller radioactive inventory simplifies some aspects of safety but complicates the long-term logistics of waste management and site security.

The realistic window for commercial deployment

The timeline for SMRs is often described in optimistic terms, but the technical and economic realities suggest a more measured pace. The first movers in the industry are currently working through the prototype and first-of-a-kind (FOAK) stages. These initial units will be expensive, likely costing significantly more than gas or renewable alternatives on a per-megawatt basis. The goal of these projects is not immediate profit but the validation of the manufacturing process and the regulatory pathway.

The realistic window for the first wave of commercial SMRs to begin contributing to the grid is the mid-2030s. Before this can happen, several milestones must be met. First, the FOAK plants currently under development in the US, Canada, and the UK must complete their construction and testing phases without the catastrophic cost overruns that defined the last generation of nuclear builds. Second, manufacturers must secure a backlog of orders that justifies the investment in high-volume production facilities. Roughly fifty to one hundred units of a single design are likely needed to reach the point where costs become competitive with other forms of low-carbon baseload power.

It is established that the engineering of small reactors is sound; several designs are based on proven light-water technology used in naval propulsion for decades. What is contested is whether the cost savings from factory production can actually outweigh the lost economies of scale inherent in small cores. The picture would change significantly if a major industrial power committed to a large-scale, state-backed purchase of a single design, providing the volume necessary to kickstart the learning curve. Without such a commitment, SMRs risk remaining a niche technology, confined to remote locations or specific industrial applications where the high cost of energy is less of a deterrent. The transition from a series of successful demonstrations to a global industry is a matter of market coordination and sustained capital, rather than any further breakthrough in nuclear physics.