About 25 miles west of Knoxville, Tennessee, was a sparsely populated 60,000 acres of land near the Blackoak Ridge. Blackoak runs north-south and connects two bends of the Clinch River, and it is part of a sequence of five ridge/valleys on the southeast side of the Appalachian Mountain Range. The Cherokees claimed it as a hunting ground, but by 1800 the Treaty of Holston had ceded it to the United States and several farming communities took root in the area.
In 1902 the local mystic, John Hendrix, 37 years old and thought by some to be not right in the head, was enjoying a typical day by lying in the woods on the ground clutter and gazing up at the sky through the trees. His attention was grabbed by a loud voice, telling him to remain there asleep for 40 nights so that he could be shown visions of what was in store for the surrounding acreage. Being given an account of the future by an external source, he repeated this information many times to anyone who would listen. His predictions were positively eerie:
And I tell you, Bear Creek Valley someday will be filled with great buildings and factories, and they will help toward winning the greatest war that ever will be. And there will be a city on Black Oak Ridge and the center of authority will be on a spot middle-way between Sevier Tadlock’s farm and Joe Pryatt’s place. A railroad spur will branch off the main L&N line, run down toward Robertsville and then branch off and turn toward Scarborough. Big engines will dig big ditches, and thousands of people will be running to and fro. They will be building things, and there will be a great noise and confusion and the Earth will shake. I’ve seen it. It’s coming.
Hendrix went on to inhabit a mental institution, and in October 1942 Brigadier General Leslie Groves, head of the Manhattan Engineer District and assigned the task of developing an atomic bomb, chose a spot between the Tadlock and Pryatt farms in east Tennessee as his headquarters.[33] It was remote, cut off from the world, and yet blessed with a great deal of surplus electrical power. The Tennessee Valley Authority, set up by President Roosevelt as a make-work project in the throes of the Great Depression, had gotten a little too enthusiastic and peppered all of east Tennessee with hydroelectric plants. Groves would quickly put them to good use.
Nobody among the Axis Powers that were trying to take over the world had ever heard of the place, and it wasn’t even on a map. It was perfect for top-secret work. In a couple of years, it would be known as the Manhattan District HQ, the Clinton Engineering Works, or simply as Oak Ridge, and its population would explode into 75,000 people. Tracks were laid for a rail spur off the L&N, right where Hendricks had said, and in Bear Creek Valley was erected an enormous industrial complex of seven buildings comprising the Y-12 site. The largest building in the world, the half-mile-long K-25 gaseous diffusion plant, was built at a bend in Poplar Creek, about 10 miles southwest of Y-12.[34] Construction began at a furious pace, making an instant city. Housing for workers, thoughtfully made of asbestos to prevent fires, was a priority. Eventually the city had ten schools, seven theaters, 17 restaurants, 13 supermarkets, a library, a symphony orchestra, churches, and its own Fuller Brush salesman.
This bustling metropolis was built from scratch for exactly one purpose: to take mined uranium, which was nearly all worthless uranium-238, and purify it down to the rare and precious ingredient, uranium-235. The atomic weight of this isotope, 235, was an odd number, and that made its heavily overloaded nucleus touchy and likely to explode if a random neutron were to blunder into it. Just any neutron would do, but it was particularly sensitive to slow neutrons, beaten down to move no faster than any common molecule at room temperature. At Y-12, K-25, and S-50 various concentrations and chemical forms of uranium-235 were stored, moved, stacked up, bottled, boxed, and formed into piles. Only the few top administrators and some of the on-site scientists knew what the stuff was for and had a vague sense of the ultimate danger of working with it. There were 12,000 workers in the K-25 building alone, and none of them was made aware of exactly what they were doing.
For the needs of military security, there was nothing better than absolute ignorance. It was impossible for a worker to spill the beans to an Axis spy, even on purpose, and this massive, continent-wide industrial effort to build atomic bombs remained unknown to the enemy powers. However, there were dangers in this enterprise that had never before visited the human race. If one happened to stack up enough of this weird material in one place, it would start to generate heat, and this energy release would increase exponentially until the stack lost its initial configuration. The conditions under which this disaster could happen were varied across multiple dimensions. The “critical mass” condition depended on the purity of the uranium-235 in the material. Uranium fresh out the ground had only 0.73 percent of the active isotope in it, and an infinite stack of it would not approach the energy production threshold. Start increasing the percentage to, say, 3.0 percent, and the probability changed. If the uranium were dissolved in ordinary water, as it often was in the stages of processing it, then the hydrogen in the water would slow down the trigger particles, the free-range neutrons, using collision dynamics. Just like a high-speed neutron hitting a hydrogen atom in water, if you crash your car into one parked, your car stops cold and the one you hit bounces excitedly into whatever is in front of it. In the same way, a high-speed neutron crashing into a slow-moving hydrogen nucleus, which is of similar mass, will kill the speed of the incoming particle. Having uranium dissolved in water, even if it’s only slightly enriched, makes a runaway fission situation quite possible.[35]
Another factor is the shape of the stack. The less surface area your stack has per volume of the stack, the better is the probability of causing an energy-release incident. The worst you can do is to stack bottles of enriched uranium oxide dissolved in water, which is a curious green color, in a rounded mound on the floor. In that configuration the surface area is minimized, so fewer neutrons, which bounce around in completely random directions, are likely to escape the stack without causing a fission. Next worse is a neat cube. The best way to stack it is in a straight, single-file line. The same number of bottles can either be benign containers of mineral water or a glowing inferno, depending on how you stack it.
Bricks of pure uranium metal are another matter of concern. Power-producing fission is possible using high-speed neutrons as triggers, freshly minted in the fission process, but the probability is lower. In pure metal, neutrons were not slowed down to desirable speed just by running into atoms. Hitting a uranium nucleus was like running your car at high speed into the side of a bank building. You might make the building move, but not by much, and your car bounces off in the opposite direction with most of its initial speed. It takes more enriched uranium mass in pure metal form to make it go nuclear than in a water solution, but that’s not to say it does not happen. Stack up enough enriched uranium metal in a shape that will encourage fission, and you have it melting through the floor.
33
The first written account of the John Hendrix story I can find is in Robinson, George O.,
34
K-25, using the gaseous diffusion process, was used to “enrich” uranium for bombs, research reactors, submarines, and power plants for the next 40 years. The other two methods for uranium enrichment, the thermo-columns at site S-50 and the electromagnetic calutrons at site Y-12, were torn down quickly after the war ended. Two large diffusion plants were built in Paducah, Kentucky, and Portsmouth, Ohio, to increase production during the Cold War, and these were copies of the Oak Ridge facilities. However, K-25 did not add to the highly enriched uranium used in the Little Boy bomb dropped on Hiroshima. The diffusion process was slow, and the uranium fed in at the mouth of the process did not have time to reach the end-product stage by the time the war ended. Using the highly inefficient Y-12 process, by August 1945 we had just enough uranium-235 for exactly one bomb.
35
Actually, uranium oxide dissolves in water, and not pure uranium metal. Another possibility is to have natural, out-of-the-ground uranium oxide dissolved in heavy water, or deuterium oxide. Using ordinary water, the plain hydrogen in it can occasionally absorb a neutron, and this is a neutron that misses the opportunity to trigger a fission. Deuterium, which is hydrogen that already has that neutron in the nucleus, doesn’t absorb neutrons, and for that reason heavy water encourages fission more than ordinary water. You could stack up bottles of plain uranium in heavy water and cause a meltdown. The dynamics of fission are that sensitive.