How Small Modular Reactors Work and Why They Matter
Small modular reactors (SMRs) promise to reinvent nuclear power — cheaper to build, safer by design, and flexible enough to power remote communities or AI data centers. Here's how they work and what stands in the way.
A New Blueprint for Nuclear Power
For decades, nuclear power meant one thing: massive, billion-dollar plants that took fifteen years to build and required a small army to operate. That model is changing. Small modular reactors (SMRs) — a new class of nuclear technology — are designed to be factory-built, shipped in modules, and installed almost anywhere. Proponents say they could finally deliver on nuclear energy's long-delayed promise: clean, reliable, around-the-clock electricity at a competitive price.
What Exactly Is an SMR?
According to the International Atomic Energy Agency (IAEA), SMRs are nuclear reactors with a power output of up to 300 megawatts of electricity (MWe) per unit — roughly one-third the capacity of a conventional reactor. The "modular" part is key: their systems and components are designed to be factory-assembled and transported to a site, rather than custom-built in place.
Like all fission reactors, SMRs generate heat by splitting uranium atoms in a controlled chain reaction. That heat turns water into steam, which spins a turbine to produce electricity. What changes is the scale, the engineering, and — in some newer designs — the coolant itself.
How the Technology Works
Most SMR designs rely on passive safety systems: instead of powered pumps and active controls to prevent meltdown, they use gravity, convection, and the basic physics of the reactor to cool themselves down automatically. This is a fundamental departure from older designs and dramatically reduces the risk of accidents, according to the U.S. Department of Energy.
Beyond light-water SMRs (which use ordinary water as both coolant and moderator), several companies are pursuing more radical approaches:
- Molten-salt reactors — use liquid salt as coolant, eliminating the need for high-pressure operation. Kairos Power's Hermes 2 design has received the first U.S. construction permit for such a reactor.
- Sodium-cooled fast reactors — TerraPower's Natrium design uses liquid sodium, enabling higher efficiency and even the potential to burn some nuclear waste as fuel.
- Microreactors — ultra-compact units under 10 MWe, intended for military bases, remote mining sites, or disaster relief.
Who Is Building Them?
The global race to commercialize SMRs is intensifying. China's Linglong One — a 125-MWe reactor in Hainan province — is expected to become the world's first commercial land-based SMR, capable of powering roughly 526,000 households. In the United States, TerraPower has broken ground at a retiring coal plant in Wyoming, and Kairos Power is building a demonstration reactor in Tennessee. In Europe, the UK has committed £280 million to Rolls-Royce's 470-MWe SMR design, while Canada's Ontario Power Generation approved construction of a GE Hitachi reactor at its Darlington site. MIT Technology Review named next-gen nuclear one of its 10 breakthrough technologies for 2026.
The Promise: Why Advocates Are Excited
The traditional nuclear industry's Achilles heel has always been cost overruns and construction delays. SMR advocates argue that factory manufacturing solves both problems. Build the same design dozens of times and costs fall sharply — the same logic that made aircraft and automobiles affordable. Smaller upfront capital requirements also make SMRs easier to finance, and their modular nature means operators can add capacity incrementally rather than betting billions on a single enormous project.
There is also a strategic angle. Tech giants including Google have signed power-purchase agreements with SMR developers to supply clean electricity for energy-hungry AI data centers — a market that conventional renewables struggle to serve reliably.
The Concerns: What Critics Point Out
The enthusiasm is not universal. A study published in the Proceedings of the National Academy of Sciences found that some SMR designs would generate significantly more radioactive waste per unit of energy than conventional plants, due to greater neutron leakage from smaller reactor cores. Critics also note that economies of scale work both ways: smaller reactors lose the cost advantages that make large plants competitive, and no SMR design outside China has yet demonstrated it can be built on budget and on time at commercial scale.
The World Nuclear Association acknowledges that the economic case for SMRs depends heavily on achieving serial production volumes that do not yet exist.
The Bottom Line
Small modular reactors represent nuclear energy's most credible attempt to reinvent itself for the 21st century. The physics is proven; the engineering is advancing rapidly; and the political will — driven by climate targets and surging electricity demand from AI — has never been stronger. Whether the economics will follow is the defining question of the next decade.
