The Nuclear Files

Sunflower rafts at Chernobyl worked on contaminated water, not soil, yet the popular story blurred the two. Measurements from Fukushima rice fields show why the difference matters for any plant-based cleanup of cesium.

A cancer-therapy machine sold as scrap for $10 in 1983 was melted into rebar and table legs, exposing about 4,000 people. It came to light only when a truckload of radioactive steel took a wrong turn at Los Alamos.

West Valley’s reprocessing plant ran about six years, and demolishing its main building took nearly three. Its vitrified waste sits on a storage pad with no disposal facility, and the final cleanup decision is still open, with 2010 estimates from $0.5 billion more to $8.2 billion.

A tiny cesium-137 capsule lost in a quarry ended up inside an apartment wall in Kramatorsk, Ukraine. Residents fell ill and died for years, and accounts still disagree on how many deaths and illnesses the source caused.

Hanford’s 1949 Green Run released iodine-131 from fuel cooled only 16 days, yet who proposed and approved the experiment remains publicly unresolved.

A radiotherapy source melted in a Spanish steel mill in 1998, and the mill’s own detectors missed it. Air filters in Switzerland, France and Italy caught the cesium first, setting off a hunt for its origin.

A plutonium-238 generator left on Nanda Devi in 1965 was never found, at the head of a watershed that feeds the Ganges. The records show a real loss, a 1978 scare and decades of testing with no documented release.

Petitioners in 2015 and an executive order in 2026 both pushed the NRC to abandon the linear no-threshold model. A proposed rule would now drop ALARA, yet the low-dose end of the risk curve is still drawn the same way.