Nuclear Fission
Six and a half years separated the discovery of fission from the destruction of Hiroshima. Otto Hahn and Fritz Strassmann's bench experiment in December 1938 found barium in a uranium target — a result so unexpected Hahn wrote to Lise Meitner asking her to explain it. She did, over Christmas walking in the snow with her nephew Otto Frisch: the uranium nucleus was splitting. By August 1945, the same physics had killed roughly 100,000 people in two cities.
How it works
A U-235 nucleus absorbs a stray neutron, becomes briefly U-236, and breaks into two unequal fragments — commonly barium and krypton, or strontium and xenon — plus 2-3 free neutrons and about 200 MeV of energy. A typical chemical bond releases 1-10 eV. Fission is roughly 20 million times more energetic per event than burning the same atom.
The energy comes from the binding-energy curve. Iron-56 sits at the minimum; everything heavier releases energy when split, everything lighter when fused. Uranium is far up the heavy side, so the fragments together weigh about 0.1% less than the original nucleus. That missing mass is the energy, via E = mc².
A fully fissioned kilogram of U-235 releases roughly 20 kilotons TNT-equivalent — the size of the Hiroshima bomb, which contained 64 kg of U-235 but actually fissioned only about 1.4% of it.
The chain reaction
Each fission spits out 2-3 neutrons. The neutron multiplication factor k decides the system's fate:
- k < 1 — subcritical. Dies out.
- k = 1 — critical. Sustains at constant rate. The reactor operating point.
- k > 1 — supercritical. Grows exponentially. The bomb operating point.
Reaching criticality needs three things: enough fissile material packed close enough (the critical mass — about 52 kg for a bare U-235 sphere, less with a beryllium reflector), neutrons slow enough to be absorbed (water or graphite as moderator in most reactors), and not too many parasitic absorbers around (the inverse of which is the control rod, boron or cadmium).
How reactors work
Fuel rods of uranium oxide, enriched to 3-5% U-235, sit in a core. Water (or heavy water in CANDU designs) both moderates the neutrons and carries away heat. Control rods slide in to absorb neutrons and damp the reaction; they slide out to let it run. The hot coolant boils a secondary water loop, which spins a steam turbine.
A reactor is a very expensive way to boil water. Everything downstream of the heat exchanger is identical to a coal plant.
The three accidents
- Three Mile Island, 1979. Partial core meltdown in Pennsylvania. Zero measurable external health impact. Killed US reactor construction for thirty years.
- Chernobyl, 1986. RBMK reactor in Ukraine pushed into an unstable configuration during a safety test. Steam explosion, graphite fire, prompt criticality. Roughly 30 direct deaths; cancer-attributable deaths estimated in the low thousands. The reactor design was uniquely dangerous and no Western reactor of similar type was ever built.
- Fukushima Daiichi, 2011. Tōhoku earthquake and 14-meter tsunami knocked out backup generators. Three reactor cores melted. Direct radiation deaths near zero; evacuation-stress deaths in the low thousands. Triggered Germany's full nuclear exit.
Nuclear's per-TWh death rate — including these three accidents — is lower than coal, oil, gas, biomass, and rooftop solar (the last mostly from installer falls). The order-of-magnitude gap between actual mortality and public risk perception is the dominant political fact about fission.
India's path
Homi Bhabha designed India's three-stage program in the 1950s: natural-uranium heavy-water reactors first, plutonium breeders second, thorium-fed reactors third. The logic was to leverage India's roughly 25% share of global thorium reserves while sidestepping its scarce uranium. First weapons test 1974 (Smiling Buddha); declared status 1998. Around 24 reactors operate as of 2026, producing about 3% of national electricity. The third-stage Advanced Heavy Water Reactor remains in design phase fifty years after Bhabha's death. The strategic logic is sound; the execution has been slower than any planner imagined.
What's contested
Whether nuclear's mortality advantage can ever overcome public risk perception is an open question, not a solved one. Cost is the harder issue: Western reactors have grown more expensive per MW over fifty years, the opposite of every other energy technology. Whether that reflects regulation, lost institutional knowledge, or genuine engineering limits is unresolved — Korea and China have built reactors on time and budget, France did so until the 1990s, and nobody fully agrees why the US and UK cannot. Small modular reactors are the current bet that factory production breaks the cost curve. No SMR has yet operated commercially at scale.
Why this has to do with other realms
The chain reaction is the cleanest physical example of an idea that recurs across concept fermi paradox and epidemiology: a multiplication factor crossing 1.0 is the difference between nothing and everything. R₀ in disease, k in fission, growth rate in concept monetary debasement — all share the same exponential mathematics and the same property that the interesting behavior is concentrated at the boundary. The hardest engineering problem in a reactor is holding k exactly at 1.000 for years. The hardest problem in pandemic response is the same number on a different process.
An open question
If SMRs deliver on factory-cost economics in the 2030s, does India skip directly to them and shelve the thorium third stage Bhabha designed? The strategic case for thorium was uranium scarcity; the strategic case for SMRs is speed.
Key sources
- The Making of the Atomic Bomb — Richard Rhodes (1986). The definitive narrative of fission's discovery through Hiroshima.
- Atomic Awakening — James Mahaffey (2009). Civilian nuclear history with engineering detail.
- Our World in Data, "What are the safest and cleanest sources of energy?" — Hannah Ritchie. The mortality-per-TWh dataset.
- To verify: IAEA Power Reactor Information System (PRIS) for current global reactor counts.
- To verify: Department of Atomic Energy (India) annual reports for Indian reactor status and AHWR timeline.
Further reading
- Command and Control by Eric Schlosser — the near-miss history of US nuclear weapons handling, which is more terrifying than the deployment history.
- Midnight in Chernobyl by Adam Higginbotham (2019) — the operational decisions and reactor physics behind the 1986 disaster.
- The Prize by Daniel Yergin — to understand why fission's commercial trajectory is best read alongside oil's.
- Whole Earth Discipline by Stewart Brand — the case for nuclear from an environmentalist who changed his mind.
See Also
- concept nuclear fusion — fission's mirror image on the binding-energy curve, with the opposite engineering problem.
- concept fermi paradox — another domain where a multiplication factor crossing 1.0 changes everything.
- overview isro cost engineering — India's other large-scale strategic technology program, and a useful contrast in execution speed.
- concept monetary debasement — uranium-backed currency was seriously proposed in the 1940s and never adopted; the reasons illuminate what makes a monetary anchor work.