Producing Plutonium

Nobel Prize-winning chemist Glenn Seaborg and physicist John Wheeler discuss the challenge and accomplishment of producing plutonium on a massive scale.

Narrator: In 1940 in a small laboratory at the University of California at Berkeley, chemists Glenn Seaborg, Joseph Kennedy and Arthur Wahl discovered microscopic quantities of a new element: plutonium. Glenn Seaborg worked on scaling up from micro-quantities of plutonium produced at Berkeley to Hanford’s huge reactors, capable of producing a billion times more plutonium for an atomic bomb.

Glenn Seaborg: There was a scale up from these micro test tubes under the microscope to the actual plant at Hanford, Washington of about a billion—a billion fold from what you might call the pilot plant to the final plant. This is the biggest scale up in the history of chemical engineering.

A lot was at stake. There were those in the laboratory who just said that this would be a huge boondoggle and would probably be Seaborg’s folly or something like this as a monument after the war. But, we stuck to our guns and insisted that the work would be valid.

John Wheeler: I never felt it [the B Reactor] as simply a massive plutonium factory. There was a sense of adventure about it that I don’t associate with any normal project. I associate it with pioneering rather than with a steady factory. The phrase “factory” conveys this sense of something that’s going on and on. “Ever wert, art, and evermore shall be.”

I would think it’s more like the early days of the first steamship. That must have been exciting. The first airplane was exciting. The first locomotive was exciting. I think of it in that category.

Ray Genereaux, a Design Engineer for DuPont, had to be creative when it came to designing a massive, first-of-a-kind chemical separations building in Hanford, Washington.

Narrator: The T Plant at Hanford was the first chemical processing and separations plant of its kind in the world. When construction began in 1943, Project leaders still had not decided what chemical process they would use to separate plutonium from the irradiated uranium. Ray Genereaux, the chief engineer tasked with designing the massive separations facility, devised an innovative solution that would accommodate multiple processes.

Ray Genereaux: They didn’t know what chemical process they would use, wet, dry, you name it. We realized immediately that we had to design flexibility into this thing if we were going to move ahead and get designs done before we knew what the hell we were doing, before we knew what the process would be. It’s quite different to design something for a wet process than it is for a dry process.

We soon realized that we had to design something that could take any kind of a process. What this meant was that we had to standardize it on the setting for equipment, which we then called a cell. Eventually, these buildings had something like forty cells in each one, and they are all designed exactly the same. There were all different kinds of connections for liquids, for electricity, for steam, for air or lubrication or instruments. That was one of the keys of the whole thing, was the standardization of the cells, so that we could put any kind of a pot or pan in them to do a job.

DuPont engineer Crawford Greenewalt and physicist Leona Woods Marshall discuss the challenge of creating water-tight aluminum cans to encapsulate fuel rods that would be inserted into the nuclear reactor.

Narrator: A little recognized problem was fitting an aluminum jacket around the fuel rods that would be absolutely water and air tight. Earl Swensson, a production manager, who took a statistical approach to finding a solution.

Crawford Greenewalt: This was very troublesome. You had thousands of these things to put in cans. The process had to be perfect enough so that you could put a thousand of them into cans and not have to reject more than a small percentage. In the first place uranium is scarce and very expensive and you couldn’t have any imperfections. It was a very big problem and it actually was not solved until we were actually out there. It was solved just in the nick of time.

Leona Woods Marshall: Nothing was going to work if those slugs didn’t work. It was one of the technicians and they, I must say, did not get enough credit. It was a clever person and I hope he got a bonus, but I doubt it.

Narrator: Earl Swensson was this clever hero. He took a statistical approach to solving the problem. The first day he made one thousand cans and they all failed the test. But ten were better than the rest and they became the models for the next day’s attempt. And so it continued until three weeks, they had 20,000 cans. Behold, there was one perfect can among them; a purely statistical process and result. After a while out of every thousand there would be 500 or 600 a day that were acceptable. And that’s how they did it.