The Screwdriver Slipped. The Room Filled with Blue Light.
Seven men watched Louis Slotin close two hemispheres around a plutonium core. When his hand slipped, he understood before anyone else.
On May 21, 1946, inside a Los Alamos laboratory, Louis Slotin held a screwdriver in his left hand. He was not using it to tighten a screw. It was keeping two beryllium hemispheres slightly apart around a ball of plutonium.
Seven men stood near the table. Some took measurements; others watched the demonstration. Slotin was showing them how to bring the material close to its critical threshold: near enough to count the neutrons, never close enough to let the reaction run away.
The upper hemisphere descended, millimetre by millimetre. The screwdriver slipped.
The lid fell. Blue light flashed across the room and a wave of heat struck Slotin’s skin. He needed neither a meter nor a calculation to understand. He tore the upper hemisphere away with his bare hand and stopped the reaction.
The laboratory fell silent again. Nothing had exploded. No wall had collapsed. Yet something had just passed through all eight men.
Slotin knew the plutonium sphere well enough to know what awaited him. It had killed once already.
The Core Had Been Made for a Third Bomb
By the summer of 1945, the Manhattan Project had already produced the weapons that would strike Hiroshima and Nagasaki. Another plutonium core was ready. If the war continued, it could become the active heart of another bomb.
Japan announced its surrender before the core was used. The sphere, weighing a little over six kilograms, remained at Los Alamos, where physicists used it to study criticality.
A plutonium core does not have to explode to become lethal. Every fission releases neutrons. If they escape, the reaction dies out. If surrounding material reflects them back into the plutonium, more atoms split, then more again. At a certain point, the multiplication sustains itself.
Researchers needed to measure that boundary because a bomb must cross it at precisely the right instant—but never during assembly, transport or storage.
In 1945, these experiments were still conducted within inches of the human body. Reflecting materials were added around the core by hand; researchers watched the counter climb, then stopped before the limit.
On paper, there was a safety margin. On the table, that margin weighed several kilograms and could fall.
Harry Daghlian Had Already Dropped the Final Brick
On the evening of August 21, 1945, physicist Harry Daghlian was working around the same sphere. He was gradually building an enclosure of tungsten-carbide bricks. Each block reflected more neutrons back towards the plutonium.
The detectors warned him: the next brick would bring the assembly too close to criticality. Daghlian moved to take it away, but it slipped from his hand and fell into the centre of the apparatus.
The reaction began. He first tried to knock the brick away, then had to dismantle part of the enclosure by hand to stop it. The blocks were intact. Daghlian had just absorbed a massive dose of neutron and gamma radiation.
He was taken to hospital. Burns appeared, his condition deteriorated and he fell into a coma. He died twenty-five days after the accident, aged twenty-four.
The rules changed: experiments were no longer to be conducted alone, and remotely operated equipment began to be designed. But the new installations were not ready yet.
The plutonium core went back to the laboratory.
Slotin Called It “Tickling the Dragon’s Tail”
Louis Slotin was one of the specialists capable of assembling the components of a bomb. After the war, he wanted to leave Los Alamos and return to teaching, but first he had to pass his knowledge on to the people who would replace him.
His experiment with the beryllium hemispheres was direct, quick and spectacular. The lower half held the core. The upper half was suspended by a handle. A spacer was supposed to prevent the two shells from closing completely.
Slotin preferred his screwdriver. He slid it between the rims and adjusted the gap with his wrist while the counters tracked the multiplication of neutrons.
Enrico Fermi reportedly warned members of the team that if they carried on this way, they would be dead within a year. The phrase used around these experiments captured their relationship with risk rather neatly: “tickling the dragon’s tail.”
On May 21, 1946, Slotin was not attempting something secret for the first time. He was demonstrating a familiar move in front of colleagues. That is precisely what makes the scene so disturbing: the danger had not been ignored. Habit had made it feel tame.
When the screwdriver slipped, Slotin immediately lifted the shell away. That reflex reduced the exposure of the other men. But his own body was directly above the reaction.
Witnesses described a blue flash or glow and a sensation of heat. The US National Park Service attributes it to the ionisation of the air caused by the intense burst of radiation. The colour did not mean the plutonium was burning. It meant that, for an instant, the entire room had become part of the accident.
The First Symptom Came Before He Reached the Door
Slotin felt a sour taste in his mouth and a burning sensation in his hand. He vomited soon after leaving the building. The men were taken to hospital, where doctors tried to estimate each person’s dose from his position around the table.
For several witnesses, the exposure was serious but not immediately fatal. For Slotin, there was almost no uncertainty.
During the first few days, he remained conscious and received visitors. Then the invisible injuries inflicted in the instant of the flash took over. His hands swelled and blistered. His digestive system ceased to function normally. Internal burns, confusion and the progressive failure of his organs gave him no respite.
He died on May 30, nine days after the accident. He was thirty-five.
The speed with which he tore the hemisphere away probably spared his colleagues a much higher dose. It does not turn the demonstration into a planned act of sacrifice: until the screwdriver slipped, everyone still believed it would hold.
After a Second Death, Distance Finally Became a Rule
Los Alamos banned manual procedures of this kind. Criticality experiments were moved to Pajarito Canyon and operated remotely. Personnel now watched from a building roughly a quarter of a mile from the apparatus.
The change did not follow any sudden discovery. Scientists already knew that radiation was unforgiving. They already knew how to calculate the threshold they were approaching. After Daghlian’s death, they had even begun to imagine machines that could do the work in their place.
What was missing was not knowledge of the danger, but the decision to make the gesture that could trigger it physically impossible.
The sphere later acquired the nickname “demon core.” The name makes the two accidents easy to recast as a curse attached to an object. It almost lets us forget that the sphere never moved by itself, never chose the final brick and never replaced a spacer with a screwdriver.
It Had Taken Two Deaths to Measure the Right Distance
The Daghlian and Slotin accidents became landmarks in nuclear safety. They show that a system cannot forever depend on the skill of its most experienced operator—especially when the smallest mistake takes its toll before a witness can even shout.
After Slotin, researchers no longer held the core at arm’s length. They watched it from another building.
A quarter of a mile now separated the human being from the plutonium. It had taken two deaths to measure that distance.
Discussion
Have something to add?
A detail, a disagreement or a lead worth pursuing? Tell us. Comments are reviewed before publication.