# Nuclear reactor A **nuclear reactor** is a device that starts, sustains and controls a [[Nuclear_chain_reaction|nuclear chain reaction]], and converts the [[Heat|heat]] it releases into useful [[Power_(physics)|power]]. Its defining trick is not that it makes energy but that it can be steered: on [[Neutron|prompt]] neutrons alone a thermal system one part in a thousand above critical would multiply its power twenty-two thousandfold in a second, which no operator and no rod drive could follow.[^derived-nr][^lamarsh-nr] What makes a reactor operable is that about 0.65 % of fission neutrons are not prompt at all. They are emitted seconds to minutes later by decaying fission products, and holding the reactor in the narrow band where those [[Delayed_neutron|delayed neutrons]] are *needed* to reach criticality stretches its response from ten-thousandths of a second to minutes. In the microsim below the reader turns one control — reactivity ρ, calibrated in cents and dollars of the delayed fraction β — and can hit a scram button. Behind it runs one-delayed-group point kinetics, `dn/dt = ((ρ − β)/Λ)·n + λ·C` and `dC/dt = (β/Λ)·n − λ·C`, with the standard pressurised-water values β = 0.0065, λ = 0.08 s⁻¹ and Λ = 10⁻⁴ s.[^spec-e49][^lamarsh-nr] The readouts are the power trace, the reactor period, and a red line at ρ = β. At $0.10 the power jumps 11 % at once and then climbs on a 113-second period; at $0.50 the period is 12.5 s; at $1.05 — past the line, prompt critical — it collapses to 0.23 s, five hundred times faster, because the delayed neutrons are no longer needed and the system reverts to Λ.[^derived-nr] The pair is stiff: at ρ = 0 its eigenvalues are 0 and −65 s⁻¹, a spread of nearly three decades, so the sim advances it with the exact 2×2 matrix exponential and never with a fixed-step integrator.[^spec-e49][^derived-nr] On the [[Energy]] flagship this article serves the *Reactor control: k-eff and delayed neutrons* section of Part V — Transformation. It is the sibling of the [[Nuclear_chain_reaction]] sim, which shows the same multiplication with k imposed and no delayed neutrons at all, and it hands the fuel forward to [[Nuclear_fuel_cycle]]. ## Terminology The fissioning region is the *core*; the *moderator* slows fast neutrons to thermal speeds where ²³⁵U fissions far more readily; the *coolant* carries the heat out, and in a light-water reactor the same water does both jobs. *Reactivity* is the fractional departure from criticality, `ρ = (k_eff − 1)/k_eff`, and because the scale that matters is set by the delayed fraction it is quoted in *dollars*, ρ/β, and cents — so $1 of reactivity is exactly prompt critical.[^lamarsh-nr] The *reactor period* is the time for power to change by a factor of e, the reciprocal of the growth rate. A *scram* is a rapid full insertion of the [[Control_rod|control rods]]. *Burnup*, in megawatt-days per tonne, measures how much energy has been taken from a charge of [[Nuclear_fuel|fuel]]. A *reactivity coefficient* is the derivative of ρ with respect to an operating variable — fuel temperature, moderator temperature, coolant void — and its sign decides whether a disturbance damps itself or grows.[^lamarsh-nr] ## Operation A reactor is four coupled systems: a chain reaction, a heat source, a coolant circuit and a [[Thermodynamic_cycle|thermodynamic cycle]] — with a control system wrapped around all of them. The coupling runs both ways: the coolant is usually also the moderator, so any change in the heat balance is immediately a change in reactivity. ### Fission Thermal neutrons absorbed in ²³⁵[[Uranium|U]] cause [[Nuclear_fission|fission]], releasing 172.3 MeV and two or three further neutrons apiece.[^murphy-nr] Those neutrons are born fast; the moderator slows them by elastic collisions with light nuclei, because a thermal neutron is far more likely to fission ²³⁵U than a fast one, and a lattice geometry that lets neutrons thermalise in the moderator before returning to the fuel is what allows a 3–5 % [[Enriched_uranium|enriched]] core to reach k_eff = 1 at all.[^murphy-nr][^lamarsh-nr] ### Heat generation Nearly all of the 172.3 MeV appears within a millimetre of the event, as the kinetic energy of the two [[Fission_product_yield|fragments]] stopping in the fuel; the rest arrives as [[Beta_decay|beta]] and gamma radiation, and a few percent leaves the plant entirely with the neutrinos.[^murphy-nr] At 172.3 MeV per fission, a 3 GW-thermal core runs 1.1×10²⁰ fissions a second and consumes about 3.7 kg of ²³⁵U a day.[^derived-nr] The heat does not stop when the chain does: fission products keep decaying, and the [[Decay_heat|decay heat]] immediately after shutdown is several percent of full power, falling roughly as t^(−0.2) in the Way–Wigner form — about 6 % at one second, 1.2 % after an hour and 0.6 % after a day.[^waywigner][^derived-nr] A 3 GW core therefore still needs 35 MW of cooling an hour after a perfect scram, which is the single fact behind most of reactor safety engineering. ### Cooling Cooling is the doublet balance `Q = ṁ·c_p·ΔT` with the pressure chosen so the coolant does not boil where it should not. A pressurised-water reactor holds its primary circuit near 155 bar so that water at over 300 °C stays liquid; with c_p ≈ 5.5 kJ/(kg·K) at those conditions and a 35 K rise across the core, removing 3 GW takes about 15,600 kg/s of flow.[^wna-nrc][^derived-nr] A boiling-water reactor lets the water boil in the core instead and sends the steam straight to the turbine. Gas-cooled designs use carbon dioxide or helium at high temperature and low [[Density|density]], and fast reactors use liquid [[Sodium|sodium]], which moderates almost not at all.[^wna-nrc] The choice is never only thermal. Because the water is also the moderator, losing it in a light-water reactor shuts the chain reaction down even as it removes the cooling; a gas or sodium coolant does not, so those designs must buy their shutdown margin elsewhere.[^wna-nrc][^lamarsh-nr] ### Reactivity control Reactivity control is the sim's subject. A rod of strong neutron absorber changes k_eff and therefore ρ; soluble boron in the coolant trims it slowly, and burnable absorbers in the fuel offset the excess reactivity of a fresh charge. What the point-kinetics equations show is that the *same* rod motion behaves completely differently on the two sides of ρ = β. Below prompt critical, a step of reactivity produces a prompt jump followed by a stable period. The jump is `n/n₀ = β/(β − ρ)`: 11 % at $0.10, a factor of two at $0.50.[^derived-nr] The period that follows comes from `ρ = Λ·ω + β·ω/(ω + λ)`, and because Λω is negligible there, the period is governed entirely by λ — 113 s at $0.10, 12.5 s at $0.50.[^derived-nr][^lamarsh-nr] Above $1 the delayed term is no longer needed, ω ≈ (ρ − β)/Λ, and the period is set by Λ alone: 0.23 s at $1.05.[^derived-nr] Operating procedures therefore limit rod worth and withdrawal rate so the prompt-critical line is never approached — which is what the sim's red line is for. Two feedbacks help. Heating the fuel broadens the ²³⁸U capture resonances and adds negative reactivity within milliseconds, faster than any rod; heating or voiding the moderator changes it too, negatively in a light-water reactor and positively in a graphite-moderated, water-cooled design — which is a property of the lattice, not of the operators.[^lamarsh-nr][^wna-nrc] Over hours, ¹³⁵[[Xenon|Xe]] builds up from ¹³⁵I decay and absorbs neutrons strongly, so a reactor shut down from full power is harder to restart for some hours afterwards.[^lamarsh-nr] ### Electrical power generation The heat drives a conventional steam plant, so the conversion obeys the same [[Carnot_cycle|Carnot]] ceiling as any other [[Heat_engine|heat engine]], `η = 1 − T_L/T_H`.[^yan-nr] A water-cooled reactor cannot deliver steam much above 300 °C without pressures that the vessel cannot carry, where a fossil boiler runs at 300–600 °C, so nuclear plants sit lower on the efficiency ladder: Murphy puts fission plants at about one-third, and a 2.5 GW-thermal to 1 GW-electric station at 40 %.[^murphy-nr][^yan-nr] Raising T_H is the main thermodynamic argument for gas-cooled, sodium-cooled and molten-salt designs. ### Lifetimes Reactors are licensed for decades and then relicensed. The limiting component is usually the [[Steel|steel]] pressure vessel, whose ductile-to-brittle transition temperature rises with accumulated fast-neutron fluence, so vessel embrittlement — not the fuel, which is replaced, nor the turbine, which is rebuilt — sets the outer bound.[^wna-nrc] Against that long life the fleet's [[Capacity_factor|capacity factor]] is high: Murphy's 2019 figures give 295 GW of average output from 393 GW installed, about 75 %, while the United States row of the same table gives 92 from 97, about 95 %.[^murphy-nr] ## History The first self-sustaining chain reaction ran on 2 December 1942 in a graphite-and-uranium pile in Chicago.[^rhodes-nr] Electricity followed at the end of 1951, when an experimental fast breeder in Idaho lit four light bulbs; the first reactor to supply an electricity grid started at Obninsk in the Soviet Union in June 1954, and the first full-scale commercial stations followed at Calder Hall in Britain in 1956 and Shippingport in the United States in 1957.[^iaea-hist] Every one was a different concept — graphite pile, fast breeder, graphite-moderated water-cooled, gas-cooled, pressurised water — and the later commercial convergence on light water owed more to naval propulsion heritage than to physics.[^wna-nrc] ### Table by date The milestone machines, in order, each the first of something: | Year | Reactor | Country | First of | |---|---|---|---| | 1942 | Chicago Pile-1 | United States | self-sustaining chain reaction | | 1951 | EBR-I | United States | electricity from fission | | 1954 | Obninsk AM-1 | Soviet Union | power to a grid | | 1956 | Calder Hall | United Kingdom | full-scale commercial station | | 1957 | Shippingport | United States | commercial pressurised water | The dates and attributions are as recorded by the international agency inventories rather than by any Portal Book.[^iaea-hist][^rhodes-nr] ### Table by country The fleet is concentrated. In 2019 it comprised 455 plants and 393 GW of installed capacity worldwide, producing about 11 % of world electricity; the United States alone accounted for 97 GW of that installed total and 92 GW of average output.[^murphy-nr] France, whose fleet is almost entirely pressurised-water, takes the largest national share of its own electricity from fission, and China has been the fastest-growing builder since the 2010s.[^wna-nrc] The pattern follows policy and industrial capacity, not resource endowment: [[Uranium|uranium]] is cheap, transportable and a small part of the cost of the electricity. ## Reactor types Designs are distinguished by three choices — moderator, coolant and neutron spectrum — and by what they are for; Kerlin's survey chapter walks the same taxonomy alongside the plant systems that hang off each choice.[^kerlin-nr] The three are not independent — a moderator that is also the coolant fixes much of the rest — which is why the commercial fleet is far less varied than the list of possibilities. ### Classifications The primary split is thermal against fast. A thermal reactor has a moderator, needs only slightly enriched fuel, and has a large ratio of moderator to fuel; a fast reactor has none, needs fuel enriched to fifteen or twenty percent or plutonium, and can breed more [[Fissile_material|fissile material]] than it consumes. Within thermal reactors the moderator may be light water, heavy water or [[Graphite|graphite]], and the coolant water, gas or [[Sodium|sodium]]. Purpose divides them again: power, propulsion, research, isotope production, and materials testing.[^lamarsh-nr][^wna-nrc] ### Current technologies Most operating power reactors are [[Pressurized_water_reactor|pressurised water reactors]], in which the primary water stays liquid and gives up its heat to a secondary circuit in a steam generator; the next largest family is the [[Boiling_water_reactor|boiling water reactor]], which boils in the core and has one circuit instead of two. Heavy-water designs use natural uranium, trading enrichment for a costlier moderator and a large core. Gas-cooled reactors, graphite-moderated and carbon-dioxide-cooled, remain in service in Britain, and a graphite-moderated water-cooled line persists in the former Soviet Union.[^wna-nrc] ### Future and developing technologies Development runs along three lines. Small modular reactors trade economies of scale for factory production and simpler passive cooling. Fast designs — sodium-cooled, lead-cooled, or gas-cooled — aim to close the fuel cycle by breeding ²³⁹Pu from ²³⁸U or ²³³U from [[Thorium|thorium]], as the [[Breeder_reactor|breeder]] and [[Thorium_fuel_cycle|thorium cycle]] pages describe. Molten-salt reactors dissolve the fuel in the coolant, which raises T_H and removes the fuel-melting failure mode by starting from a liquid.[^wna-nrc][^murphy-nr] ## Nuclear fuel cycle The fuel arrives as ceramic uranium dioxide pellets stacked in [[Zirconium|zirconium]]-alloy tubes and assembled into bundles, and it leaves as [[Spent_nuclear_fuel|spent fuel]] containing unburned ²³⁵U, bred plutonium, and the fission products that make it intensely radioactive. Whether the plutonium is recovered, and what is done with the rest, belongs to [[Nuclear_fuel_cycle|the fuel cycle]]; here it is enough that ⁹⁰Sr and ¹³⁷Cs dominate the hazard from about five years to a few hundred.[^murphy-nr] ### Fueling of nuclear reactors Light-water reactors are refuelled in batches: the plant shuts down every 12 to 24 months and roughly a third of the core is replaced, the remaining assemblies being shuffled toward positions of lower [[Neutron|neutron]] flux. Heavy-water designs refuel on load, one channel at a time, without stopping. The excess reactivity a fresh core must carry to reach the next outage is exactly what the boron and burnable absorbers suppress, so refuelling interval, enrichment and control system are one design problem.[^wna-nrc][^lamarsh-nr] ## Nuclear safety Safety engineering follows from decay heat. Because a scrammed core still produces megawatts, the requirement is not to stop the reaction — the rods do that in seconds — but to keep removing heat afterwards, through any plausible combination of failures. The standard answer is defence in depth: independent barriers between fuel and public, independent cooling paths, and increasingly passive paths needing no electricity and no operator. The negative fuel-temperature coefficient is the innermost defence, acting in milliseconds and impossible to disconnect.[^lamarsh-nr][^wna-nrc] ## Nuclear accidents The severe accidents on record share one mechanism and differ in their causes. In each, cooling was lost while decay heat continued, the fuel overheated, the zirconium cladding reacted with steam to produce hydrogen, and radioactive material escaped its barriers. One 1979 accident in the United States damaged a core without significant off-site release; the 1986 accident in the Soviet Union combined a positive void coefficient with a prohibited test and destroyed the core and building; the 2011 accident in Japan followed a tsunami that removed all cooling power from three reactors that had already shut down correctly.[^wna-nrc] What none of them was, and what Murphy is careful to state, is a nuclear explosion: reactor fuel can melt, but its enrichment and geometry cannot produce a bomb.[^murphy-nr] The three sequences defeated three different defences: the operators' picture of the plant in 1979, the reactivity coefficient itself in 1986, and every cooling path at once in 2011 — which is why passive cooling, needing neither power nor an operator, is the direction later designs have taken.[^wna-nrc] ## Natural nuclear reactors Chain reactions ran without engineers at Oklo in Gabon about 1.7 billion years ago, when natural uranium was about 2.9 % ²³⁵U — light-water reactor fuel — and ordinary groundwater served as the moderator.[^oklo-nr][^derived-nr] The deposits appear to have cycled: as the reaction heated the rock the water boiled away, moderation was lost and the reaction stopped until the water returned — a void coefficient doing exactly what one is supposed to do. ## Emissions A reactor emits no combustion products, so its operating [[Energy_transformation|carbon]] output is nil and its whole-life total comes from mining, enrichment, construction and decommissioning; on that accounting fission sits with wind and hydroelectricity rather than with fossil generation.[^theis-nr] Routine radiological releases are small and monitored. What dominates public and regulatory attention is not the routine emission but the possible one, which is why the safety case rather than the emissions inventory decides whether a reactor is built.[^theis-nr][^wna-nrc] ## See also - [[Nuclear_chain_reaction]] — the sibling sim, k imposed and no delayed neutrons - [[Delayed_neutron]] — the 0.65 % that makes control possible - [[Control_rod]] - [[Pressurized_water_reactor]] - [[Boiling_water_reactor]] - [[Nuclear_meltdown]] - [[Decay_heat]] — the after-scram tail in Way–Wigner form - [[Nuclear_fuel_cycle]] - [[Nuclear_power]] ## Notes The sim's kinetics model uses **one** delayed group. Real fission products fall into six conventional groups with decay constants from about 0.012 to 3 s⁻¹, and collapsing them to a single λ = 0.08 s⁻¹ reproduces the stable period well near critical while compressing the long tail after a scram, where the longest-lived precursors dominate. The decay-heat curve quoted here is the Way–Wigner correlation, a fit to measurements rather than a derived law, and is ILLUSTRATIVE at very short and very long times. Reactivity in dollars is a ratio, ρ/β, so a dollar is a different absolute reactivity in every fuel: β is about 0.0065 for ²³⁵U and roughly a third of that for ²³⁹Pu, which makes a plutonium-fuelled core proportionally harder to control. Footnote definitions for the whole page are collected under *References*, below. ## References [^murphy-nr]: Murphy, Thomas (2021). *Energy and Human Ambitions on a Finite Planet*. Chapter 15: Table 15.7 and the 172.3 MeV per fission, pp. 273–274; the 2019 fleet at 455 plants, 393 GW installed and 295 GW average, 11 % of world electricity, with the United States row at 92/97 GW, p. 276, and the cost comparison assuming ~90 % capacity factor, pp. 276–277; fission plants at about one-third efficiency and a 2.5 GW-thermal to 1 GW-electric plant at 40 %, p. 276; control rods described in principle with no equation, p. 275; the ⁹⁰Sr and ¹³⁷Cs waste window, p. 281; a core can melt but cannot explode like a bomb, p. 282. https://open.umn.edu/opentextbooks/textbooks/energy-and-human-ambitions-on-a-finite-planet [^yan-nr]: Yan, Claire Yu (2022). *Introduction to Engineering Thermodynamics*. Chapter 6: the Carnot efficiency `η = 1 − T_L/T_H` and the impossibility of exceeding it, p. 272; large steam plants raising T_H with 300–600 °C steam, p. 273. The equation displays were lost in extraction and are supplied here in standard form. https://open.umn.edu/opentextbooks/textbooks/introduction-to-engineering-thermodynamics [^kerlin-nr]: Kerlin, Thomas (2013). *Future Energy: Opportunities & Challenges*. Chapter 8, pp. 267–353 (page to pin): reactor types, plant systems, fuel, safety and accidents. Chapter 3, p. 185, for the rule that a thermal plant rejects 2.5–3 times its electrical output. https://open.umn.edu/opentextbooks/textbooks/future-energy-opportunities-challenges [^theis-nr]: Theis, Tom; Tomkin, Jonathan, eds. (2015). *Sustainability: A Comprehensive Foundation*. Chapter 10, *Sustainable Energy Systems* (page to pin): life-cycle emissions accounting and the comparison of generating technologies. https://open.umn.edu/opentextbooks/textbooks/sustainability-a-comprehensive-foundation [^lamarsh-nr]: Lamarsh, John R.; Baratta, Anthony J. *Introduction to Nuclear Engineering*, 3rd ed. Chapters on reactor theory, reactor kinetics and reactor licensing (page to pin): the point-kinetics equations `dn/dt = ((ρ − β)/Λ)·n + λ·C` and `dC/dt = (β/Λ)·n − λ·C`; the delayed fraction β ≈ 0.0065 for ²³⁵U and the six-group decay constants; reactivity in dollars and cents; the inhour relation `ρ = Λ·ω + β·ω/(ω + λ)`; the prompt jump `β/(β − ρ)`; the prompt neutron lifetime of order 10⁻⁴ s in thermal systems; the fuel-temperature (Doppler) and moderator coefficients; and ¹³⁵Xe poisoning from ¹³⁵I decay. **No Energy Portal Book states reactor kinetics** — Murphy gives the chain and the control rods in principle only, p. 275 — so every kinetics result on this page is cited to this standard text rather than assigned a Portal Book page. [^waywigner]: Way, Katharine; Wigner, Eugene P. (1948). "The Rate of Decay of Fission Products." *Physical Review* (page to pin). The `P/P₀ ≈ 0.0622·(t^(−0.2) − (t + t₀)^(−0.2))` correlation used here is the standard form derived from that work and is reproduced in [^lamarsh-nr]; it is an empirical fit, not a derived law. [^wna-nrc]: World Nuclear Association technology briefings and United States Nuclear Regulatory Commission technical reference material (page to pin): pressurised- and boiling-water primary conditions near 155 bar and above 300 °C, reactor-type inventories and national fleets, refuelling practice, vessel embrittlement as the life-limiting mechanism, defence in depth and passive safety, and the accident sequences at Three Mile Island (1979), Chernobyl (1986) and Fukushima Daiichi (2011). No Energy Portal Book carries these at the level of detail used here. [^iaea-hist]: International Atomic Energy Agency, Power Reactor Information System and associated historical inventories (page to pin): EBR-I (1951), Obninsk AM-1 (1954), Calder Hall (1956) and Shippingport (1957) as first-of-kind machines. [^rhodes-nr]: Rhodes, Richard. *The Making of the Atomic Bomb* (Simon & Schuster, 1986), page to pin: the first self-sustaining chain reaction on 2 December 1942. [^oklo-nr]: International Atomic Energy Agency, reporting on the *Oklo natural fission reactors*, Gabon (page to pin): the reactor episodes dated to about 1.7 billion years ago and their apparent cycling. No Portal Book covers Oklo. [^spec-e49]: Matter & Energy Cluster contract, `_registry/plans/ENERGY_SECTIONS.md` row E49: the sim concept, one-delayed-group point kinetics with β = 0.0065, λ = 0.08 s⁻¹ and Λ = 10⁻⁴ s; reactivity in cents and dollars as the control rod, plus a scram button; readouts of reactor period, power trace and the prompt-critical line at ρ = β; and the requirement that the system be stepped by the exact 2×2 exponential because it is stiff. [^derived-nr]: Computed for this article from the equations on the page and the cited constants: prompt-only growth `exp((k − 1)·t/ℓ)` = e¹⁰ = 2.2×10⁴ in one second at k = 1.001 and ℓ = 10⁻⁴ s; the prompt jump `β/(β − ρ)` = 1.11 at ρ = $0.10 and 2.00 at $0.50; stable periods from `ρ = Λ·ω + β·ω/(ω + λ)`, giving ω = 0.00889 s⁻¹ (T = 113 s) at $0.10, ω = 0.08 s⁻¹ (T = 12.5 s) at $0.50, and ω = 4.41 s⁻¹ (T = 0.23 s) at $1.05, a ratio of about 500 across the prompt-critical line; the kinetics matrix at ρ = 0 has trace −65.08 and determinant 0, hence eigenvalues 0 and −65.08 s⁻¹ — the stiffness that forces the closed-form step; fission rate and burn rate at 172.3 MeV = 2.76×10⁻¹¹ J, giving 1.1×10²⁰ fissions per second and 3.7 kg of ²³⁵U per day for a 3 GW-thermal core; coolant flow `ṁ = Q/(c_p·ΔT)` = 3×10⁹/(5,500 × 35) = 15,600 kg/s; Way–Wigner decay heat at 6.2 % (1 s), 1.2 % (1 h) and 0.64 % (1 d), so 35 MW an hour after a 3 GW scram; the Oklo enrichment of 2.9 % at 1.7 Ga from the half-lives of [^murphy-nr] as derived on [[Nuclear_chain_reaction]]. ## External links - [*Energy and Human Ambitions on a Finite Planet*](https://open.umn.edu/opentextbooks/textbooks/energy-and-human-ambitions-on-a-finite-planet) (Murphy, 2021), Chapter 15 — the fission energetics and the fleet figures used here - [*Future Energy: Opportunities & Challenges*](https://open.umn.edu/opentextbooks/textbooks/future-energy-opportunities-challenges) (Kerlin, 2013), Chapter 8 — reactors, plant systems and safety - [*Introduction to Engineering Thermodynamics*](https://open.umn.edu/opentextbooks/textbooks/introduction-to-engineering-thermodynamics) (Yan, 2022), Chapter 6 — the Carnot ceiling on the steam side - The Wikipedia pair's *External links* section lists the agency and regulator sites for the reactor inventories and accident reports cited above. <!-- MATTERSIM:BEGIN g33 — Matter & Energy Cluster microsim (framework build, specs/sims/Nuclear_reactor.json); do not hand-edit inside --> **Microsim — three.js (Wikitube framework):** *Nuclear reactor* <div class="wt-sim" data-src="https://wikitube-3d-microsims.netlify.app/matter/Nuclear_reactor.html" data-title="Nuclear reactor"></div> *Built from `MICROSIM_GUIDE/specs/sims/Nuclear_reactor.json`; part of the [[PORTAL_Matter|Matter portal]] spine (section sims and See-also variants).* <!-- MATTERSIM:END --> ## Wikipedia : Wikitube **Strict pair:** [Wikipedia](https://en.wikipedia.org/wiki/Nuclear_reactor) : [Wikitube](https://en.wikitube.io/wiki/Nuclear_reactor) · pinned revision [1369563173](https://en.wikipedia.org/w/index.php?oldid=1369563173) · 2026-09-11 ## Previous hub tags Hubs: `Life_Physics`. Portals: [[PORTAL_Energy]]. --- *Matter & Energy Cluster child articles, wave 1 · 2026-09-11 · drafted · Energy row E49 · sim pending (matter/Nuclear_reactor).*