Energía y Generación Eléctrica

Nuclear Reactors in 2026: what the official manual leaves out

Post actualizado el día octubre 17, 2026 by DeiviSanzPlay

Nuclear reactors don’t burn anything. They split atoms. That’s the whole trick, and it’s been working commercially since 1954, when the Obninsk plant in the Soviet Union first pushed electricity onto a grid. If you want to understand how a nuclear reactor operates, stop thinking about fire and start thinking about billiard balls. Uranium-235 atoms are the balls, neutrons are the cue, and every collision throws off more cue balls than you started with. That’s a chain reaction.

Fission: The Split That Powers Cities

When a slow-moving neutron hits a U-235 nucleus, the nucleus absorbs it, becomes unstable, and tears apart into two lighter elements plus two or three fresh neutrons. Those neutrons smash into other U-235 atoms. One split becomes two, two become four, four become eight. That’s how a nuclear reactor generates heat: pure fission, no combustion.

The energy released per split is roughly 200 MeV. Sounds abstract, so here’s the street version: one kilogram of uranium-235 fissioned completely releases about 24,000,000 kWh of heat. A coal plant needs around three million kilograms of coal to match that. That’s why utilities put up with the complexity.

Not every neutron keeps the party going. Some leak out of the core, some get eaten by non-fissile U-238, some get absorbed by structural materials. The reactor only sustains a chain reaction when you hit criticality, meaning each fission produces exactly one neutron that triggers another fission. Supercritical means power is rising. Subcritical means it’s dying. Operators live on that knife edge, deliberately.

From Heat to Steam to Kilowatts

The heat from fission gets carried away by a coolant, usually pressurized water, boiling water, or in some designs liquid metal or helium gas. In a pressurized water reactor (PWR) — the most common type worldwide — water sits at around 155 bar so it can reach 315°C without boiling. That hot water runs through a steam generator, transferring heat to a secondary loop that does boil.

That secondary steam hits a turbine. The turbine spins a generator. The generator converts mechanical energy into electricity. Same Rankine cycle a coal plant uses. The nuclear part is just a very expensive, very clean way to boil water.

After the turbine, steam gets condensed back into liquid and pumped back to the steam generator. Closed loop. Nothing radioactive touches the turbine in a PWR, which is one reason this design dominates: the radioactive water stays inside containment.

Control Rods and the Art of Not Blowing Up

Control rods made of boron, cadmium, or hafnium absorb neutrons without splitting. Push them deeper into the core, fewer neutrons survive, power drops. Pull them out, power rises. That’s your throttle. In most reactors, they’re the primary power control mechanism, though boric acid dissolved in the coolant also does fine-tuning.

Two safety layers matter more than people realize. First, prompt neutrons — the ones released instantly at fission — only account for about 99.35% of the neutron population. The rest are delayed neutrons, released seconds to minutes later from decaying fission products. That delay is why reactors are controllable at all. Without it, power would spike faster than any mechanical system could react.

Second, most modern reactors have a negative temperature coefficient. If the core gets hotter, the fission rate naturally drops. Physics does the braking for you. Chernobyl’s RBMK design had a positive coefficient at low power, which is exactly why it blew up during a botched test in 1986.

Cooling and Why «Meltdown» Is a Loaded Word

Even after a reactor shuts down, decay heat keeps generating about 7% of full thermal power, dropping to around 1% within hours. That sounds small. It’s not. At 3,000 MW thermal, 1% is 30 MW of heat with nowhere to go if pumps fail.

That’s exactly what happened at Fukushima Daiichi in 2011. The tsunami knocked out backup generators, cooling stopped, fuel cladding reacted with steam, and hydrogen explosions wrecked three reactor buildings. The cores partially melted. A «meltdown» is technically fuel damage, not a nuclear explosion. Fission reactors cannot detonate like a bomb because the fuel isn’t enriched enough — commercial fuel runs at 3% to 5% U-235; weapons need over 90%.

Modern designs address this with passive cooling. The AP1000 uses gravity-fed water tanks that need no electricity for 72 hours. Small modular reactors like NuScale’s design rely on natural circulation. That’s the direction the industry is actually heading in 2026.

Safety as Layers, Not a Checklist

Defense in depth means redundant barriers: fuel pellet matrix, zirconium cladding, steel pressure vessel, reinforced concrete containment, and the building itself. Any single barrier can fail without releasing radiation. All of them failing at once is the scenario regulators lose sleep over.

Redundancy extends to pumps, valves, and power supplies. Plants carry N+2 redundancy in critical systems, meaning you can lose two independent cooling trains and still have one working. Operators train in full-scale simulators every six weeks. They drill for station blackout, loss of coolant, and steam generator tube rupture until responses are muscle memory.

Seismic and flood design uses historical data plus margin. Japan’s Onagawa plant, closer to the 2011 epicenter than Fukushima, survived the tsunami intact because its seawall was built higher. Design choices matter more than slogans.

Carbon, Waste, and the Honest Trade-Off

Nuclear power’s lifecycle emissions run around 12 grams of CO2 per kWh, according to IPCC data. Coal sits near 820 g/kWh, natural gas around 490. That gap includes uranium mining, enrichment, and construction. It’s not zero, but it’s in the same league as wind.

High-level waste is the real problem. A typical 1,000 MW reactor produces about 27 tonnes of spent fuel per year. That waste stays dangerously radioactive for thousands of years. Finland’s Onkalo repository, expected to start operations in the mid-2020s and continue for a century, is the first permanent deep geological disposal site. Everyone else is watching.

Land use cuts the other way. A 1,000 MW nuclear plant needs about 1 to 2 square miles. An equivalent solar farm needs 50 to 75 square miles. That’s the argument nuclear advocates keep making, and it’s a fair one.

Fourth-generation reactors and fusion research promise less waste and better fuel efficiency. Fusion, in particular, would produce no long-lived actinides. But ITER won’t produce net energy before the 2030s, and commercial fusion is realistically a 2050s conversation. Until then, fission is what we have, and it works.