THERMAL DIFFUSION

• Centrifuges
• Electromagnetic Separation
• Gaseous Diffusion
• Thermal Diffusion
• Separation Hazards and Wastes

The basic principle in liquid thermal diffusion is that the lighter molecules in a liquid will concentrate in areas of higher temperature. As used at the S-50 facility, the process involved placing liquid uranium hexafluoride between two concentric vertical pipes. Uranium hexafluoride is a molecule consisting of one uranium atom and six fluorides. Fluoride only exists in one isotope, so any difference in mass among the molecules of uranium hexafluoride must come from the different isotopes of uranium. To create a temperature differential, the outer pipe was cooled and attracted the heavier uranium-238, while the inner pipe was heated and attracted the lighter uranium-235. Convection currents gradually brought the relatively more concentrated uranium-235 to the top of the pipes where it was collected. The taller the column, the greater the separation.

In early 1940, scientists evaluating the various methods for uranium isotope separation ranked thermal diffusion behind only the centrifuge and gaseous diffusion. Initial efforts focused on gaseous thermal diffusion, but experiments at Columbia University and the University of Minnesota quickly established that the process was impractical for large-scale separation. Philip Abelson proposed instead using liquid thermal diffusion and, with support from the Navy, began experimentation first at the Carnegie Institution, then at the Bureau of Standards, and finally, in June 1941, at the Naval Research Laboratory in Washington, D.C., where higher steam pressures and superior shops were available. Using 36-foot vertical columns, Abelson began tests with uranium hexafluoride, a compound so little known that he developed his own method for producing it in quantity. Initial results were disappointing, as Abelson found that the degree of separation involved both spacing between the hot and cold walls and hot wall temperature, but by summer 1942 he had increased the uranium-235 concentration of hexafluoride by up to 21 percent. Vannevar Bush and James Conant received reports about Abelson's research but concluded that even to increase the uranium-235 concentration by 50 percent would take too long for the thermal diffusion process to make a major contribution to the bomb effort, especially since the electromagnetic and pile projects appeared more promising. General Leslie Groves visited Abelson's laboratory in September 1942 but discounted the project because it appeared to lack a sense of urgency. Improvements in the process, with Abelson now using 48-foot columns, prompted Bush and other officials to visit the laboratory in January 1943. A thorough review determined that to construct a plant producing one kilogram of fully enriched uranium-235 per day would require 18 months and $75 million. The simplicity of the plant, with no moving parts and no valves in the hexafluoride system, made it attractive, but the long start-up times assured that the plant would not be in full production until early 1946. Manhattan Project officials therefore recommended only further research and preliminary engineering studies. Abelson continued his work with the Navy independently of the Manhattan Project, obtaining authorization to build a new 100-column pilot plant at the Philadelphia Naval Yard, where construction began in January 1944.

In early 1944, J. Robert Oppenheimer learned about Abelson's progress in thermal diffusion research. Faced with problems in the K-25 gaseous diffusion plant and in the Y-12 electromagnetic separation plant, Oppenheimer informed Groves in late April that development of thermal diffusion could provide slightly enriched uranium feed to the Y-12 plant. A review ordered by Groves found that if the K-25 powerhouse was used to provide steam a 1600-column plant—capable of enriching 50 kilograms of uranium a week to slightly less than 0.9 percent concentration—could be built quickly at a cost of $3.5 million. On June 24, Groves approved construction of the S-50 liquid thermal diffusion plant immediately adjacent to the K-25 powerhouse in Oak Ridge.

When completed, the S-50 plant consisted of 2,142 columns, each 48 feet in height, distributed in 21 racks. Each column had 3 concentric tubes, with a 1¼-inch nickel pipe inside, a slightly larger copper pipe in the middle, and a 4-inch galvanized iron jacket on the outside. The one-hundredth of an inch annular space between the nickel pipe's outer wall and the copper pipe's inner wall contained the uranium hexafluoride. Steam, under a pressure of 100 pounds per square inch and at a temperature of 285 degrees Celsius, circulated downward through the nickel pipe. Water at 68 degrees Celsius flowed upward through the iron jacket. Uranium hexafluoride flowed into the base of each column from a reservoir. At the top of each column was a system of freezing coils that permitted small amounts of enriched product to be drawn off at frequent intervals. The 102-column racks occupied a single main process building, a large black structure 522 feet long, 82 feet wide, and 75 feet high. Running the length of the west side was a mezzanine with 11 control rooms, one for every two racks, and an equal number of transfer rooms containing process equipment for both supplying feed material and removing enriched product and depleted uranium hexafluoride from the columns.

The S-50 process produced an "enriched" mixture in which the percentage of uranium-235 was raised from 0.71 percent to about 0.852 percent. Part of the S-50 plant was operating in September 1944, only 70 days after the start of construction, but initial production was meager with only 10.5 pounds produced in October, 171.8 pounds in November, and 20 pounds in December. With 6 racks in operation by the first of the year, production in February 1945 climbed to 3,158 pounds and, with the full plant in operation, to 12,730 pounds in June. The slightly enriched uranium was sent to the Y-12 electromagnetic plant for further enrichment and then, beginning in late April 1945, to the K-25 gaseous diffusion plant. The total S-50 output accelerated by about a week the flow of weapons-grade uranium-235 from the Y-12 Beta tracks to the Los Alamos laboratory.

• Centrifuges
• Electromagnetic Separation
• Gaseous Diffusion
• Thermal Diffusion
• Separation Hazards and Wastes

Previous      Next