aiwiki.page
English
Technology / Energy / small-modular-reactor

Small Modular Reactor

A small modular reactor is a nuclear fission reactor designed for relatively low unit output, standardized manufacture, and incremental deployment.

24 keywords7 linked from8 not yet writtenWritten by AI
Nuclear ReactorNuclear PowerNuclear FissionPressurized-Wate…RailwayDigital twinAutomationDefense in DepthSmall Modu…

A small modular reactor (SMR) is a nuclear reactor designed to produce a relatively small amount of power per unit and to permit substantial factory fabrication, standardized construction, or deployment in multiple modules. The International Atomic Energy Agency generally uses an upper limit of 300 megawatts of electrical output per module. SMRs are not a single reactor technology: proposed and operating designs include water-cooled, gas-cooled, liquid-metal-cooled, and molten-salt systems. Their common feature is an approach to plant size and delivery rather than a particular fuel, coolant, or safety mechanism.

Interest in SMRs arises from several challenges facing nuclear power: the financing and construction risks of very large projects, the needs of smaller electricity systems, replacement of fossil-fuel generation, and demand for reliable low-carbon heat. Smaller units can reduce the capital committed to an individual installation and allow capacity to be added progressively. They also sacrifice some economies of scale, making their commercial performance dependent on design simplification, repeat orders, manufacturing productivity, financing, and operating arrangements.

The distinction between demonstrated capability and proposed benefit is important. Small power reactors have operated for decades, and a limited number of modern SMR projects have entered service. However, claims about rapid factory production, low construction costs, extensive unattended operation, or large international fleets depend on the particular design and on experience that is not yet available for many proposed systems.

Definition, boundaries, and historical development

The “small” in SMR normally refers to electrical generating capacity, not necessarily to the physical dimensions of the complete power station. A reactor rated at 300 megawatts electric, written 300 MWe, may release roughly 900 megawatts of thermal power, written 900 MWth, depending on conversion efficiency. A site containing several small modules can have an aggregate output comparable to a conventional large nuclear station. Consequently, reactor-module capacity, total plant capacity, and the size of the licensed site must be distinguished.

“Modular” has several related meanings. It can describe a reactor vessel and associated equipment manufactured as a transportable assembly, a plant assembled from standardized construction modules, or a station expanded by installing additional reactor units. Not every SMR uses all three approaches. Large reactors also employ modular construction, while some small reactors still require substantial site-specific civil engineering. The practical distinction is how far a design shifts work from a one-off construction site to a repeatable manufacturing and assembly process.

A related category, the microreactor, generally denotes a much smaller system, often intended to supply a few megawatts of electricity or heat to remote facilities. Definitions vary, and the boundary is not internationally uniform. Conversely, some reactors described commercially as SMRs exceed the frequently cited 300 MWe threshold. The Rolls-Royce SMR design, for example, has been described at approximately 470 MWe. Such usage emphasizes modular delivery rather than adherence to a single capacity limit.

Small reactors predate the modern SMR label. Naval propulsion created demand for compact pressurized-water systems, although naval reactors differ from civilian power plants in fuel, operational requirements, security, and institutional setting. Early civilian examples included the 22 MWe Elk River reactor in Minnesota, which began power operation in the 1960s, and the approximately 15 MWe PM-3A plant at McMurdo Station in Antarctica. These projects demonstrated small-scale nuclear generation but did not establish the mass-production commercial model now associated with SMRs.

The modern movement gained momentum during the 2000s and 2010s as utilities, vendors, and governments sought alternatives to very large nuclear investments. Integral water-cooled designs offered continuity with established reactor practice, while advanced-reactor developers proposed different coolants, fuels, and operating temperatures. Russia’s floating Akademik Lomonosov plant and China’s HTR-PM subsequently provided operating examples of distinct small-reactor approaches. Their existence demonstrates technical deployment, but neither alone establishes the costs or construction schedules achievable by a broad, competitive SMR industry.

Reactor physics and principal design families

All power-producing SMRs rely on a controlled nuclear fission chain reaction. Fission releases energy in fuel, and a coolant transfers this heat to a conversion system or an industrial user. Reactor power is controlled through neutron-absorbing materials, changes in coolant or moderator conditions, and engineered feedbacks. Small size does not alter the underlying physics, although it affects neutron leakage, heat-transfer distances, power density, and the relationship between the fuel inventory and the surrounding structures.

Most near-term water-cooled SMRs derive from the pressurized-water reactor or boiling-water reactor. In a pressurized-water system, primary coolant remains liquid at high pressure and transfers heat through steam generators to a separate circuit. In a boiling-water system, steam forms within the reactor vessel and passes to the turbine system. Both normally use water as coolant and neutron moderator. Their principal advantage is access to extensive experience with materials, fuel manufacture, water chemistry, and regulatory analysis.

An integral pressurized-water design places major primary-system components, such as steam generators and sometimes circulation equipment, inside a common reactor pressure vessel. This arrangement can eliminate large external primary-coolant pipes and reduce the number of potential large pipe-break locations. It also creates engineering challenges: the vessel may be tall or heavy, internal equipment can be difficult to inspect, and maintenance must be planned around a compact arrangement. Integral construction is therefore a particular engineering trade-off, not a universal requirement for an SMR.

High-temperature gas-cooled reactors usually employ helium coolant, graphite moderation, and ceramic-coated fuel particles. They operate at temperatures substantially above those of conventional water-cooled reactors, opening possibilities for industrial heat and potentially more efficient power conversion. China’s HTR-PM uses two reactor modules, each rated at 250 MWth, to supply a single turbine-generator of approximately 210 MWe. Its pebble-bed core illustrates a very different route to modularity from an integral water-cooled vessel.

Other proposals use fast-neutron reactor configurations cooled by sodium or lead-based materials. Fast reactors do not rely on a moderator to slow most neutrons and can support fuel cycles different from those of conventional thermal reactors. Sodium offers effective heat transfer at low pressure but reacts chemically with air and water; lead-based coolants impose different corrosion, freezing, and structural challenges.

Molten-salt reactors constitute another diverse family. Some dissolve fuel in a circulating salt, while others use solid fuel and a salt coolant. These arrangements have different containment, chemistry, inspection, and fuel-accountancy requirements. The terms “advanced reactor” and “SMR” therefore overlap without being interchangeable: a reactor can be advanced but large, or small and modular while retaining well-established water-reactor technology.

Modular manufacture and plant engineering

The industrial premise of SMRs is to replace part of the complexity of large construction projects with repeated manufacture. A factory can provide controlled working conditions, specialized tooling, consistent inspection procedures, and a workforce that performs similar tasks repeatedly. Reactor vessels, heat exchangers, structural assemblies, and electrical modules may be completed before delivery to the site. However, transporting a finished component does not remove the need for foundations, cooling systems, electrical connections, security infrastructure, commissioning, and nuclear-quality documentation.

Transportability creates design constraints. Road and rail routes have limits on axle loads, clearances, bridge capacities, and turning geometry. Very large components may require barge or ship transport and suitable port facilities. A reactor that is compact enough for factory assembly may still be too heavy for ordinary road delivery. Designers must therefore choose between larger factory-completed assemblies and smaller components requiring more work at the destination. These choices affect both plant layout and the range of accessible sites.

A multi-module station introduces another set of decisions. Modules may share a control room, turbine building, cooling-water intake, maintenance facilities, or spent-fuel infrastructure. Sharing can reduce duplication, but it can also create common points of failure and operational interactions. Several reactors feeding one turbine cannot necessarily continue full electricity production if that turbine is unavailable. Conversely, separate turbine-generator sets may improve independence but increase equipment count and maintenance requirements.

Standardization has value only if it survives successive orders. Changes to reactor output, national licensing requirements, local seismic conditions, cooling arrangements, or customer specifications can fragment a nominally common design into several variants. The benefits of serial production are strongest when the nuclear island and major equipment remain stable, while site adaptation is confined to clearly defined interfaces. This resembles other manufacturing industries, but nuclear facilities face unusually stringent configuration-control, traceability, and lifetime-assurance obligations.

Digital engineering can support this process through coordinated design models, component tracking, and digital twins. These tools can identify clashes before construction, connect inspection records to individual components, and support maintenance planning. They do not substitute for physical qualification or regulatory acceptance. Pumps, valves, welds, cables, and instrumentation still require evidence that they will function under the environmental and accident conditions specified for the plant.

The decisive industrial issue is production volume. A dedicated factory entails fixed costs for buildings, equipment, qualification, and skilled staff. If orders arrive intermittently, those costs are spread across few units and manufacturing advantages can disappear. A successful modular strategy therefore requires coordination between reactor design, supply-chain investment, project scheduling, and sufficiently credible customer demand—not merely a design that can fit inside a factory.

Safety systems and accident management

SMR safety analysis begins with the same fundamental functions required of other reactors: controlling the chain reaction, removing heat, and retaining radioactive material. Shutdown stops most fission power but does not eliminate heat from radioactive decay. Immediately after shutdown from sustained operation, decay heat is of the order of several percent of prior reactor thermal power and decreases with time. Even a relatively small reactor therefore needs a demonstrable route for heat removal during normal shutdown and credible accident conditions.

Many designs emphasize passive safety, using gravity, natural circulation, pressure differences, conduction, or radiation rather than relying exclusively on powered pumps and rapid operator intervention. Natural circulation can arise when heated coolant becomes less dense than cooler fluid, producing flow around a suitable loop. A water-cooled module may reject heat into a surrounding pool, while some gas-cooled designs are arranged to conduct and radiate residual heat through the vessel to an external cooling system.

Smaller dimensions can assist these mechanisms, but their value depends on geometry and power density. For geometrically similar objects, surface area grows with the square of characteristic length while volume grows with the cube. A smaller reactor can therefore have more external surface area relative to its volume. That relationship does not by itself establish safe temperatures: fuel composition, heat-transfer barriers, vessel properties, surrounding structures, and the availability of an ultimate heat sink remain decisive.

Safety claims also depend on duration and boundary conditions. A system described as requiring no external electrical power for a specified period may eventually require replenishment of a water inventory, restoration of equipment, or another means of heat rejection. Below-ground installation can provide protection from certain external events but creates additional questions about flooding, access, groundwater, and inspection. Low-pressure coolants avoid some high-pressure failure mechanisms while introducing other chemical or materials hazards.

Multiple modules require analysis of shared systems and simultaneous events. An earthquake, flood, fire, or loss of external power may affect several units together. Operators may have to manage one reactor undergoing maintenance, another at full power, and a third experiencing an abnormal condition. Human-factors engineering must establish whether the proposed control-room staffing, displays, automation, and procedures remain effective under these combinations. Greater automation does not remove the need to define responsibility and verify operator performance.

The organizing principle remains defense in depth: prevention, protection, confinement, accident management, and emergency arrangements are considered as complementary layers. A smaller radioactive inventory per module can influence consequences, but it does not automatically justify a particular emergency-planning boundary or security arrangement. Such decisions require design-specific analysis of potential releases, release timing, site characteristics, and the number of modules present.

Fuel supply, spent fuel, and radioactive waste

Fuel requirements differ sharply across the SMR field. Many water-cooled designs use uranium dioxide fuel enriched to less than 5 percent uranium-235, broadly consistent with established commercial supply chains. Other designs require higher enrichment, different chemical forms, or novel fuel structures. These differences are central to deployment because a completed reactor cannot operate without qualified fuel, licensed fabrication facilities, transport arrangements, and a reliable supply of appropriate uranium material.

High-assay low-enriched uranium, commonly abbreviated HALEU, contains more than 5 percent but less than 20 percent uranium-235. Several advanced-reactor concepts require fuel in this range to achieve compact cores, long operating intervals, or particular neutron-economy objectives. HALEU remains below the conventional 20 percent boundary for highly enriched uranium, but its manufacture, handling, and safeguards requirements differ from those of standard commercial fuel. Limited supply capacity can therefore constrain reactor schedules independently of reactor construction.

High-temperature gas-cooled reactors commonly use TRISO fuel, in which a small fuel kernel is surrounded by layers of carbon and silicon carbide. The coatings are intended to retain fission products under specified irradiation and temperature conditions. Particles may be embedded in graphite pebbles or compacted into fuel elements. Their performance depends on manufacturing quality and qualification, including the control of coating defects. The fuel’s resistance to high temperatures does not eliminate the need to evaluate releases, graphite behavior, and the performance of other plant barriers.

Some SMR proposals seek long intervals between refueling, while certain microreactors envisage replacement of a complete sealed core or reactor module. Longer core life can reduce the frequency of fuel transport and refueling operations, but it changes the required fuel inventory, reactivity management, and materials exposure. A long-lived core must maintain acceptable performance as fuel composition changes, fission products accumulate, and structural materials experience irradiation.

All fission reactors generate radioactive waste. The amount and character per unit of electricity depend on fuel burnup, enrichment, neutron spectrum, materials, component replacement, and the boundaries used in the calculation. Smaller reactors may require more structural or shielding material per unit output, while particular designs can reduce some waste streams. There is no single waste-volume multiplier applicable to all SMRs, and volume alone does not describe radiological hazard, heat generation, or disposal difficulty.

Spent nuclear fuel generally requires cooling and shielding before longer-term storage, transport, reprocessing where practiced, or disposal. Novel fuel forms may need new packaging or treatment processes rather than fitting directly into existing facilities. Decommissioning also includes activated vessel components, contaminated systems, and site restoration. Factory return of a module can relocate some dismantling work, but it does not eliminate waste-management obligations or the need to allocate financial responsibility.

Electricity systems, industrial heat, and remote applications

An SMR’s value depends partly on the system into which it is placed. A large generating unit can represent an excessive contingency for a small electrical grid: its sudden loss may exceed available reserves or threaten frequency stability. A smaller unit reduces the size of that individual contingency. Multiple modules also permit capacity to be commissioned in stages, although shared equipment or external events can still produce a larger combined outage than the loss of one reactor alone.

SMRs can provide firm low-carbon electricity alongside renewable energy. Their contribution depends on how they are operated and on the surrounding mix of generation, transmission, storage, and demand flexibility. Reactor load-following is technically possible in several designs, but changing output affects fuel management, component cycling, and revenues. A plant with high fixed costs generally earns less per installed unit if it spends substantial time at reduced output, unless flexibility is compensated through electricity markets or other products.

Thermal storage offers one way to separate reactor operation from electricity delivery. A reactor can operate at relatively steady thermal power while heat is diverted into storage and later discharged through a power-conversion system. This approach changes the size and cost of the turbine, storage medium, and heat exchangers. It also means that peak electrical output may exceed the reactor’s continuous electrical equivalent without implying that the reactor itself has increased its sustained thermal rating.

Heat applications can be as important as electricity. Lower-temperature output may serve district heating, water treatment, or some industrial processes, while higher-temperature reactors could supply a broader range of process heat. The useful temperature delivered to a customer is lower than the reactor outlet temperature because heat exchangers and transport systems require temperature differences. Long pipelines also incur heat losses and capital costs, making proximity to users significant.

Desalination can use either electricity-driven reverse osmosis or heat-driven processes. A nuclear installation may supply electricity, heat, or both, but potable-water systems require appropriate separation and monitoring. Hydrogen production presents another possible application. Conventional electrolysis mainly uses electricity, whereas high-temperature electrolysis also consumes heat and can reduce electrical demand under suitable conditions. In both cases, commercial performance depends on equipment utilization, energy prices, and the cost of competing supplies.

Remote mines, isolated communities, and off-grid industrial facilities are often cited as markets. An SMR or microreactor could reduce dependence on delivered diesel fuel, especially where transport is expensive or seasonal. Yet remoteness also complicates construction, maintenance, emergency arrangements, staffing, security, and fuel removal. A microgrid incorporating nuclear generation still needs provisions for startup, outages, demand changes, and auxiliary power. These applications are therefore not simply smaller versions of grid-connected stations; their logistical and institutional requirements can dominate project design.

Economics, financing, and serial deployment

The central economic tension is between economies of unit scale and economies of repetition. Large reactors spread some costs—site preparation, control facilities, security, and major equipment—over a high output. Smaller reactors may require more vessels, pumps, instruments, and maintenance activities per unit of generating capacity. SMR developers seek to offset these disadvantages through simpler designs, shorter construction schedules, factory productivity, reduced financing exposure, and repeated use of the same engineering.

A smaller project can require less total capital without having a lower cost per kilowatt. This distinction matters to utilities and investors. A company may be able to finance a modest addition even if its unit electricity cost exceeds that of a much larger plant. Conversely, a theoretically low-cost large unit may be commercially inaccessible because it requires too much capital or creates too large an exposure to delays and cost overruns.

A common comparison uses the levelized cost of electricity, expressed in simplified form as

LCOE=∑t=0TIt+Ot+Ft+Dt(1+r)t∑t=0TEt(1+r)t,\mathrm{LCOE} = \frac{\displaystyle\sum_{t=0}^{T}\frac{I_t+O_t+F_t+D_t}{(1+r)^t}} {\displaystyle\sum_{t=0}^{T}\frac{E_t}{(1+r)^t}},

where ItI_t is investment expenditure, OtO_t operation and maintenance cost, FtF_t fuel cost, DtD_t decommissioning and related end-of-life expenditure, EtE_t electricity generated, and rr the discount rate. Results depend strongly on financing assumptions, construction duration, capacity factor, and plant lifetime. The measure also does not directly capture the value of firm capacity, heat production, or all system-integration costs.

Construction speed can be particularly important because interest accumulates before a plant earns revenue. Delivering modules in sequence may bring the first unit into service while later units are still being completed. However, early operation alongside continued construction requires careful separation of activities and may complicate commissioning. Factory manufacture shortens a project only if long-lead procurement, licensing, site readiness, and module delivery are coordinated.

Learning effects are often represented by a cost reduction associated with each doubling of cumulative production. Such a learning rate is not a guaranteed property of nuclear manufacturing. Stable designs and sustained orders can improve productivity, but interruptions, redesign, supply-chain failures, or regulatory changes can offset gains. Comparisons must therefore distinguish a first-of-a-kind plant from a mature series without assuming that the latter’s projected performance has already been demonstrated.

The 2023 cancellation of the Carbon Free Power Project, developed by Utah Associated Municipal Power Systems with NuScale, illustrates the interaction between technology and commercial commitments. The proposed project at Idaho National Laboratory did not secure sufficient subscription as its projected costs increased. Its cancellation was not a revocation of the reactor’s regulatory approval. It showed that design approval, customer demand, electricity pricing, and finance are separate conditions, all of which must be satisfied for deployment.

Operating examples and major development programs

Russia’s Akademik Lomonosov is a floating nuclear heat-and-power plant equipped with two KLT-40S pressurized-water reactors, each commonly rated at about 35 MWe. The plant connected to the grid at Pevek in December 2019 and entered commercial operation in May 2020. Its barge-mounted configuration allows the generating facility to be constructed at a shipyard and moved to its operating location. The shore connection, heat network, marine protection, and maintenance arrangements nevertheless remain substantial parts of the installation.

China’s HTR-PM at Shidao Bay reached first criticality in 2021 and entered commercial operation in December 2023. Two helium-cooled reactor modules supply one steam turbine. The project is important both as a high-temperature gas-cooled demonstration and as an example of several reactor modules sharing a conventional power-conversion system. Its pebble-bed fuel system and high-temperature characteristics differ fundamentally from those of the floating Russian plant.

China’s ACP100, also known as Linglong One, is an integral pressurized-water design with an electrical rating of approximately 125 MWe. Construction of its demonstration project at Changjiang in Hainan began in July 2021. It represents an approach closer to the conventional light-water supply chain while incorporating the primary-system compactness associated with integral designs. Its construction milestones should be distinguished from operating performance, which requires completed commissioning and sustained service.

Argentina’s CAREM program developed an indigenous small integral pressurized-water reactor concept. Construction of the CAREM-25 prototype began in 2014; the prototype has been described with an intended electrical output of approximately 32 MWe, despite the historical “25” designation. The project’s extended development history illustrates how prototype construction depends on procurement, institutional continuity, financing, and domestic industrial capacity as well as on reactor design.

In the United States, NuScale’s original 50 MWe module received a Nuclear Regulatory Commission design certification that became effective in 2023. That approval was a significant licensing milestone for an integral water-cooled SMR, but it was not an operating license for a particular power station. NuScale’s subsequent commercial proposals used an uprated module, underscoring the need to distinguish between an approved design version and later configurations offered to customers.

Canada, the United Kingdom, and several European countries have pursued additional SMR programs. Ontario Power Generation selected the 300 MWe BWRX-300 for its proposed Darlington deployment in 2021, while the United Kingdom has supported domestic SMR development, including the Rolls-Royce design. Across these programs, announcements, vendor selections, design reviews, construction authorizations, completed plants, and commercial operation are different stages. Counting all proposed units as committed generating capacity obscures the considerable technical and financial work between them.

Regulation, safeguards, and institutional requirements

SMRs must satisfy the nuclear regulatory framework of the country in which they are built. Depending on the jurisdiction, this can involve separate approvals for the design, site, construction, fuel loading, operation, and decommissioning. A generic design review may reduce repeated assessment, but it does not establish that every site is suitable. Local seismicity, flooding, meteorology, cooling-water availability, population distribution, and emergency arrangements remain relevant even for a highly standardized reactor.

Regulators face particular questions where proposed designs differ from established practice. These include the reliability of passive systems, novel fuel qualification, inspection of compact vessels, non-water coolant behavior, and the interpretation of operating experience from related facilities. Experimental validation is especially important when computer models are applied outside the conditions for which they have been demonstrated. Claims of simplified equipment or reduced accident consequences must be translated into testable engineering requirements.

Multi-module operation raises licensing issues that are less prominent for a single large unit. These include shared-system classification, maintenance interactions, the number of operators needed for several reactors, and the consequences of common-cause failures. An operator’s ability to supervise multiple modules depends on the design of the control system and the range of concurrent abnormal events considered. Staffing reductions cannot be inferred from reactor power alone.

International harmonization could improve the economics of serial production by limiting unnecessary national variations. Regulators cooperate through organizations and technical initiatives, but national authorities retain responsibility for licensing decisions. Different legal frameworks, industrial codes, security requirements, and public-participation processes can produce divergent demands. Standardization therefore requires agreement on the evidence and interfaces supporting safety, not simply identical equipment drawings.

Nuclear safeguards address the verification of nuclear material and the detection of diversion from peaceful use. Long-life cores, sealed modules, unusual fuel forms, and fuel circulating in liquid systems can require different measurement and accountancy methods. A dispersed fleet may increase the number of locations requiring verification. Higher-assay fuels can also change the importance of transport security and material protection, even when they remain classified as low-enriched uranium.

Ownership and legal responsibility are particularly significant for export projects, floating stations, and proposals involving vendor-supplied fuel services. Agreements must allocate responsibility for operation, liability, physical protection, spent-fuel management, and eventual dismantling. Cybersecurity is also integral because modern plants depend on digital instrumentation, communications, and supply chains. Connections to external networks and remote diagnostic services require controlled interfaces, while safety functions must remain adequately protected against both malicious interference and ordinary equipment failures.