How Did Scientists Test The Demon Core At Los Alamos?

I've always been fascinated by the weird mix of hands-on tinkering and high-stakes physics those Los Alamos experiments had. The core people later called the 'demon core' was tested the old-fashioned way: you make a subcritical assembly and then you add neutron-reflecting material around it bit by bit while watching the neutron instruments. Practically that meant stacking heavy tungsten-carbide bricks or closing beryllium hemispheres around the plutonium mass to bounce escaping neutrons back into the fuel. Each added reflector raises the effective multiplication factor, so the experimenters could see how close they were to criticality by monitoring neutron count rates, oscilloscopes, and ionization chambers. The goal was to map reactivity — how the assembly responds as neutron economy changes — not to explode anything, but to understand at what point the chain reaction becomes self-sustaining.

I was struck, reading about those days, by how tactile the work was. Instead of remote rigs and interlocks we have now, scientists literally adjusted spacer shims, stacked bricks by hand, or used a screwdriver to keep two beryllium halves slightly apart. That screwdriver trick is infamous: people later called it 'tickling the dragon's tail' because it was a delicate tease toward prompt criticality. When the separation failed, the core emitted an intense burst of neutrons and gamma rays — a bluish flash and a spike on the detectors — and those nearby got severe radiation doses in an instant. Two tragedies stand out: one incident involved a dropped brick and another involved the screwdriver slipping; both resulted in fatal exposures days later. Those events shocked the community into modernizing procedures.

After those accidents the culture changed fast. Measurements that used to be eyeballed were redesigned for remote control, mechanical interlocks and automated shutters became standard, and teams moved to monitor from behind thick shielding or via instruments that could be read out at a distance. Film badges and ion chambers stayed, but shockingly banal tools like a screwdriver were swept out in favor of engineered jigs that would prevent human fingers from being the last line of defense. Reading about it now — flipping through photographs, procedural notes, and personal remembrances — you get a real sense of how a handful of mistakes reshaped experimental safety for decades. It’s a grim lesson, but one that pushed safer, smarter methods into the mainstream of experimental nuclear work, and that feels oddly reassuring.

I tend to tell this story fast when friends ask: at Los Alamos they were doing approach-to-critical experiments by bringing neutron reflectors closer to a plutonium core and watching how neutron counts changed. One method was stacking dense tungsten-carbide bricks around the subcritical assembly; another famous method used two beryllium hemispheres that could be slowly closed to reflect more neutrons back into the core. The idea is simple in principle — reflectors reduce neutron leakage, so less fissile material is needed to sustain a chain reaction — but in practice it was nerve-wracking work done almost by hand.

The safety culture then was nowhere near what we expect now. Instruments like Geiger counters, ionization chambers, and film badges monitored exposures, but the experiments were often performed with people standing nearby making tiny adjustments. Two of the best-known mishaps occurred when a brick was dropped and when a screwdriver slipped during the hemisphere test, each producing a prompt critical excursion and fatally irradiating the experimenters. After that the labs moved quickly to remote controls, engineered barriers, and strict procedural checklists. If you want a historical read that connects the human side with the physics, pick up 'The Making of the Atomic Bomb' — it’s full of these tense, human moments that show how much trial, error, and, sadly, tragedy shaped modern practices.