1940

Gaseous Diffusion

Thomas Graham (1805–1869), Francis Simon (1893–1956), Nicholas Kurti (1908–1998)

In 1848, Scottish physical chemist Thomas Graham proposed what is now known as Graham’s law: the rate at which gases escape through porous barriers varies with the square root of their mass. If you compare two pure gases, one with a molecular weight that’s four times as heavy as the other, the lighter gas will diffuse through a barrier (or escape through a very small pinhole) twice as fast as the heavier one. This was later found to be a result of the general kinetic theory of ideal gases, and it remained just another physical proof of that theory for almost one hundred years.

In the early 1940s, the scientists of the Manhattan Project were producing the first atomic bomb, and gaseous diffusion became an essential step. They had to enrich uranium to produce uranium-235, the critical isotope in an atomic fission reaction (the splitting of an atomic nucleus to release enormous quantities of energy). However, since isotopes have almost exactly the same chemical reactivity, effects like Graham’s law that depend on molecular weight had to be enlisted. Uranium is very far from being a gas, but strenuous chemistry converted it to gaseous uranium hexafluoride. The difference in weight between uranium-235 and uranium-238 isn’t much, but using multiple cascading centrifuges with semipermeable membranes allowed the gaseous isotopes to be separated.

The gaseous diffusion process was first developed in 1940 by German-born physical chemist Francis Simon and Hungarian-born physicist Nicholas Kurti, both of whom fled from Germany to England upon Adolf Hitler’s rise to power. It was perfected in Oak Ridge, Tennessee—a town that ended up using a significant fraction of all the electricity being generated in the U.S. by the end of the war. Gaseous diffusion alone still didn’t enrich the uranium enough, so the material from the centrifuges was passed through an early type of mass spectrometry apparatus to get material suitable for a bomb. But gaseous diffusion is a suitable technique to make low-enrichment uranium for nuclear power plants, and a new plant, the first in many years, is being built in the state of Ohio for just that purpose.

General Leslie Groves, commander of the Manhattan Project, speaks to a crowd at Oak Ridge, Tennessee, in 1945. The city did not even exist in early 1942, but two years later it was consuming almost 15 percent of the electric power output of the entire country, much of which ran giant centrifuges.