Taking off from Fairfield-Suisun AFB, a USAF Boeing B-29-NR loses power in its number two engine. In accord with precautionary measures of the time, the nuclear weapon it carries lacks a fissile core. The plane lifts off, but the landing gear cannot be raised, increasing drag. The pilot circles back for an emergency landing, but the heavily laden bomber does not reach the runway.

Twelve occupants are killed in the crash; the other eight receive minor injuries. Twenty minutes after the crash, during firefighting efforts, the 10,000 pounds of high explosive in the bomb detonate. Another 180 people are killed or injured. Although the bomb is destroyed, radiological contamination of the site is negligible.

In the sixth year of operation of Canada's NRX research reactor in Chalk River, a complicated series of operator errors and safety system malfunctions leads to a massive power excursion which causes partial meltdown of the core. The accident floods the basement with a million gallons of cooling water carrying 10,000 Curies of long-lived radioisotopes. Despite a total estimated yield of 1.2x10²⁰ fissions, worker exposures are apparently low.

A subsequent hydrogen explosion throws the four-ton gas-holder dome to its maximum height, 1.2m. The containment vessel is not breached. But since air is being used to cool some of the fuel rods, thousands of curies of fission products are released into the atmosphere, making this the first major reactor disaster. Lt. Jimmy Carter, then a nuclear engineer in the U.S. Navy, takes part in the cleanup. That involves pumping a million gallons of contaminated water out of the reactor building into shallow trenches near the Ottawa River, removing and burying the core, and installing a replacement. The NRX is back in service in one or two years.

Plutonium is being recovered from spent fuel rods at the Mayak Production Association in the Soviet Union. This operation involves many processing steps on the plutonium nitrate solution. On Sunday, 15 March, operators ignore several established procedures. The most significant violations are using vessels next to each other (which allows neutronic interactions) and exceeding the limit of 500 grams of solution per vessel. The operators transfer the contents of two vessels (#2 & #4) into a third (#18). Another factor is that vessel 18 already holds 5 liters of solution at the beginning of the shift — a transfer for which there is no log entry. Noticing vapor from the solution in vessel 18, and feeling its warmth, the operator there immediately reconnects the transfer hose. Some of the solution is pulled into a vacuum trap, ending the fission reaction.

The two operators, untrained in criticality safety, elect not to report the event. (Also, the facility has no criticality alarm systems.) The accident is not discovered until two days later, when the operator near vessel 18 gets sick. His dose is estimated as 1,000 rads. He loses both legs, but lives 35 more years.

The criticality measurement table at the Sarov (Arzamas-16) facility in the southern Urals is designated FKBN. Its vertical piston is hydraulically operated and incorporates no fast-acting scram mechanism.

On 9 April, during the lunch hour, a single operator is working the apparatus alone in violation of buddy-system rules. He also places steel separators half as thick as called for by the experiment plan (5mm vs. 10 mm) on the lower half of the fissile mass. As the table raises this lower half to the upper, criticality occurs. The resulting heat melts a portion of the plutonium. The excursion sets off an alarm.

About 2 hours later, the operator and a supervisor examine the apparatus closely, receiving low doses of 1-2 rads. The yield of the excursion is about 10¹⁶ fissions. The active portion of the apparatus is shipped to the Mayak facility for further analysis and disposition. The remaining parts are clean enough for immediate re-use. However, the FKBN is replaced by a model with better safety features.

During the early morning, the crew of the Lucky Dragon, a Japanese fishing boat in mid-Pacific, think they see the sun rising to the west of them. It is instead "Castle Bravo" — a 6-MT hydrogen bomb fired at Bikini Atoll, 85 miles away. Four hours later, white ash begins to fall like snow on the boat. Many crew members scoop it off the decks and into bags as souvenirs. By evening, all 23 crew members are sick. They are hospitalized in Japan, where one dies due to radiation-induced kidney failure.

The white flakes also fall on the 86 residents of Rongelap Atoll, closer to the test site, soon piling up half an inch deep. Similar fates befall the residents of progressively more distant atolls named Rongerik, Alininae, and Utirik. By the time they are evacuated 50 to 75 hours later, many are sickened. The first warning of radiation in the fallout is raised seven hours after the blast, when the ash reaches the U.S. personnel at the weather station on Rongerik. They are moved out within 30 hours. These atolls are part of the Marshall Islands, a trust territory for which the U.S. is responsible.

The incident causes a rift between Japan and the U.S., which has not warned other nations about the test. (A mitigating factor is that an overlooked reaction boosted the bomb's yield to about 15 MT and thus expanded the danger area.) The U.S. issued an apology and paid $2 million in compensation. The exclusion was widened for later tests.

At the National Reactor Testing Station near Idaho Falls, ID, the BORAX-I has been well-studied and is about to be dismantled. Its operators decide to run one more transient test: an overload which will melt about 4% of its fuel plates.

In the event, the power excursion and subsequent steam explosion completely destroy the core and rupture the tank surrounding it. Pieces of the tank are found 200 feet away. Because of the remote location, nothing else is damaged and no one is exposed to radiation.

EBR-1, the experimental breeder reactor at the National Reactor Testing Station, has operated for three years and met its objectives. A final study of "prompt positive power coefficient without coolant flow" is conducted. When power reaches 1 MW, the scram signal is given. Unfortunately the slow motor-driven system is activated by error instead of the fast-acting gravity-driven system. Later examination shows that half the core has melted and vaporized sodium-potassium eutectic has forced some of the highly reactive coolant into the reflector.

The source concludes its report with this optimistic note: "During this accident no one received more than trivial radiation from airborne fission products, and direct exposure was essentially zero." Public Citizen quotes a report that says: "The public is not made aware of this meltdown until Lewis Strauss, head of the Atomic Energy Commission, is confronted by the Wall Street Journal and has to admit his ignorance of the accident."

Two explosions destroy a portion of Sylvania Electric Products' Metallurgy Atomic Research Center in Bayside, Queens, NY. Nine workers are injured.

A CDC document indicates that this facility was engaged in research on uranium and thorium powder metallurgy, and also investigated the health effects of beryllium and uranium powders in its main building until the AEC Medical Division required it to construct a separate laboratory for health studies. According to the CDC document, referenced below, the facility operated from 1947 through 1962 or 1965, the State of New York declared the site decontaminated of radioactivity in 1985, and condominiums were subsequently built on the site.

A U.S. B-47 practising landings at Lakenheath Air Base in Suffolk, England, skids into a bunker holding three Mark VI bombs. A secret cable to SAC commander Gen. Curtis LeMay from Gen. James Walsh, commander of U.S. 7th Airborne Division, reports that "The B-47 tore apart the igloo and knocked about three Mark Sixes. Aircraft then exploded, showering burning fuel over all. Crew perished. ... Preliminary exam by bomb disposal officer says a miracle that one Mark Six with exposed detonators sheared didn't go." Gen. Walsh refers to the risk of the bombs' high-explosive charges being set off by the fire, spreading radioactivity over the area.

A Defense Department description of the accident says "no capsules of nuclear materials were in the weapons or ... the building."

Using remote-control handling equipment in an isolation box, H. E. Northway, manager of the Houston plant of M. W. Kellogg Co., opens a can containing ten pellets of Iridium-192. He and another employee are in the room. They discover that two of the pellets are "powderized" but apparently do not report this fact to the AEC. Some of the dust escapes the box and at least one of the two employees becomes contaminated. Although the company is licensed by the AEC to encapsulate sources for radiographic cameras, the agency does not learn of the problem for five weeks.

The event is reported in the 13 May 1956 issue of Time (and reportedly by Look Magazine in 1961.) By then, at least eight private homes and seven automobiles have been contaminated. Only the two employees suffer radiation injuries, but press coverage leads to ostracism of plant workers in their community and widespread public fears of radioactivity.

At the Rocky Flats weapons facility 27 km from Denver, CO, finely-divided plutonium powder in a glove box begins to spontaneously combust about 10 PM. The fire is not immediately noticed, and soon the plastic glove box ignites. The ventilation system pulls burning fragments through the building's air ducts to the paper stack filters that keep contamination inside. The filters are damaged, allowing plutonium and other contaminants to escape the plant. Firefighting efforts begin at the glove box at 10:38. The filter fire is detected later and put out by 2 AM on 12 September.

Research done in 1999 finds that approximately 63 kilograms (kg) of plutonium were in the room where the fire occurred and that 13-21 kg were actually involved in the fire. This leads to an estimate that between 40 and 500 grams of plutonium were carried off the site by wind currents. (Another source reports the amount of plutonium lost as 25,618 micrograms, but this seems suspiciously precise.)

Cooling system failure causes an explosion in a waste storage tank at the Mayak nuclear fuel complex near Chelyabinsk, Russia. The force of the explosion is estimated to equal that of 75 tons of TNT. It releases approximately 20 MCi (700 PBq) of waste, exposing (by various estimates) 124,000 to 270,000 people to dangerous levels of radiation.

This event, known as the "Kyshtym Disaster", is merely the best known of a series of releases that both precede and follow it. More than 500,000 inhabitants of the region are affected, both by direct exposure and through contaminated water supplies. It is known to the U.S. government but withheld from the public until 1977, when a Freedom of Information Act (FOIA) request breaks it free. Russia does not admit it until 1992, when Glasnost (openness) takes hold.

During an operation to release Wigner Energy from the graphite portions of Windscale Pile No.1 at Sellafield, north of Liverpool, England, technicians mistakenly overheat the reactor pile because poorly placed temperature sensors indicate it is cooling rather than heating. The resulting fire burns four days in the air-cooled pile, consuming a significant portion of it. A deluge of water finally cools the pile and quenches the blaze.

Radiation levels in the area are comparable to those from the 1979 meltdown at Three Mile Island in the U.S. Milk distribution is banned in the 200 square miles surrounding the plant, located in west Cumbria. Contamination is also detected in southeast England. The full report of the inquiry is not made public until 1988. At that time, it is disclosed that a canister of polonium 210 also burned in the fire. Both Pile No. 1 and No. 2 are later shut down and decommissioned in an effort that lasts until 1999.

A B-47 carrying a nuclear weapon, with its fissile core in a separate container for safety, crashes just after takeoff into an inhabited area 3,800 feet from the end of the runway at Homestead AFB, Florida. Enveloping the weapon and its core, the fire burns and smolders for four hours.

Afterward the fissile core is recovered intact in its shipping container. Just half the nuclear weapon remains, but all its major components are recovered in damaged condition. Nothing is reported about radiation levels in the area.

Near Savannah, Georgia, a B-47 collides with an F-86 fighter during simulated combat. The damaged bomber attempts three landings at Hunter Air Force Base, but cannot land safely with its cargo: a nuclear weapon with no plutonium core. The crew jettisons the bomb from 7,200 feet over water near Tybee Island; it is never recovered.

Note: Lutins puts the date of this accident at 25 May 1958.

Shortly after taking off from the U.S. air base at Greenham Common, England, a B-47E of the 310^th Bomb Wing develops problems and jettisons its two external 1,700-gallon fuel tanks. Both tanks miss their designated safe impact area. One hits a hangar; the other strikes the ground 65 feet behind another B-47E parked on the tarmac. The parked B-47E, carrying a B28 1.1-MT hydrogen bomb and with a pilot aboard, is engulfed by burning jet fuel. The fire ignites parts of the parked aircraft's magnesium fuselage and sets off the B28's high explosive charges. Two men are killed and eight injured. Convection currents distribute plutonium and uranium oxides over a wide area. Sixteen hours and a million gallons of water are needed to put out the fire.

The American and British governments throw a blanket of secrecy over the event. As late as 1985, the U.K. claims that a taxiing aircraft struck a parked one, and that no fire was involved. However two scientists with the Atomic Weapons Research Establishment at Aldermaston, F. H. Cripps and A. Stimson, working independently, discover high levels of radioactivity around the base in 1960. Their report on the accident remains under wraps until 1996.

In the reactor at the Boris Kidrich Institute in Vilna, Yugoslavia, a subcritical foil counting experiment is in progress. The operators desire the maximum foil activation, so the reactor is run as close to critical as possible. After about 5 minutes, one of the operators smells ozone and realizes that the reactor has gone critical. Of the three neutron flux sensors, two are indicating steady power level. The third is disconnected after behaving erratically. It is later found that the two working sensors have reached saturation, while the power level is ramping up.

The reactor reportedly withstands the burst energy of about 2.6x10¹⁸ fissions (80 MJ), but the six operators present are strongly irradiated, receiving estimated doses of 200 to 433 rem. One dies; the other five recover after severe bouts of radiation sickness.

Power output of the High Temperature Reactor Experiment (HTRE-3) at the National Reactor Testing Station is being raised to 120KW under control of a servo-system. After reaching 80% of this level, the indicated power begins to fall off. The servo pulls the control rods out farther, but the indicated power continues to drop. The reactor scrams automatically 20 seconds later. (It is thought that this is triggered by thermocouple wires melting open.)

The cause turns out to be control instrumentation failure. A newly added noise filter in the high voltage line to the ion chamber is sensitive to high neutron flux, dropping the chamber's voltage as the flux increases and thus mis-leading the servosystem. In the excursion, of about 2.5x10¹⁹ fissions, all core fuel elements experience some melting. Some fission products escape the building, but operator personnel exposures apparently are negligible.

At the Santa Susana Field Laboratory 30 miles northwest of Los Angeles, Rocketdyne operates the 20-MW Sodium Reactor Experiment (SRE) for the AEC. SRE is a prototype of a high-temperature liquid-metal-cooled reactor for civilian applications. A cooling channel becomes blocked. This causes 30% of the fuel elements in the core to melt. Most radionuclides are contained in the reactor building; only some krypton and other noble gases are released, after being held for a time to allow their activity to diminish. Of the nine radiological incidents at SSFL, this is the only one from which detectable radioactivity is observed off-site.

The Santa Susana Field Laboratory opened in 1948 and operated to circa 1989. Rocketdyne Corporation used it primarily to test rocket motors; but ten research reactors (all but SRE under 1 MW output) and seven criticality test setups were operated in Area IV. Over the years, the site became contaminated not only with radionuclides but with trichloroethylene, mercury and hydrazine. Workers were also irradiated at the site, and studies done in the 1990s document a statistical correlation between cumulative doses and cancer rates among these workers. It should be noted that weather and topographic factors limit the spread of contaminants into surrounding populated areas.

A chemical explosion occurs during decontamination of processing equipment at the radiochemical processing plant at Oak Ridge. The cause is thought to be hot nitric acid mixing with phenol left in an evaporator which, contrary to procedures, has not been flushed with water. No injuries result, but the building is extensively contaminated, and radiation escapes it to wind up on surrounding streets and the exteriors of nearby buildings. An estimated 15g quantity of plutonium-239 is lost.

Areas that cannot be decontaminated are covered with concrete or bright warning paint. Subsequently, all radiochemical processing apparatus is fitted with secondary containment.