diesel engines, batteries and fuel.
Enrichment and separation are different goals, but the same technologies achieve them. Abelson began work by looking up the record of those technologies. Gaseous barrier diffusion was under study then at Columbia, electromagnetic separation at Berkeley, centrifuge separation at the University of Virginia. Abelson decided to try a process that had been pioneered in Germany before the war: liquid thermal diffusion (using glass tubes, Otto Frisch had experimented unsuccessfully with a similar process, gaseous thermal diffusion, at Birmingham). Thermal diffusion relied on the tendency of Ughter isotopes to diffuse toward a hotter region while heavier isotopes diffused toward a colder region. The mechanism for driving such diffusion could be simple: a hot pipe inside a cold pipe with liquid uranium hexafluoride flowing between the two pipe walls. Depending on the difference in temperature and the spacing between the two pipes more or less diffusion would occur. At the same time the heating and cooling of the hex would start a convection current flowing up the hot pipe wall and down the cold. That would bring the U23 5- enriched fluid to the top of the column where it could be tapped off. To increase the enrichment a number of columns could be connected in series to make a cascade like the cascade of barrier tanks planned for K-25.
Abelson's first technical contribution, in 1941, was inventing a relatively cheap way to make uranium hexafluoride. He processed the first hundred kilograms of hex produced in the United States. For the nominal sum of one dollar the Army contracted to borrow his patented process for Oak Ridge. He never saw the dollar.
The experimental thermal-diffusion columns Abelson built at the Naval Research Laboratory in 1941 and 1942 were 36 feet tall, each consisting of three pipes arranged one inside the other. The hot inner pipe, 11/4 inches in diameter, carried high-pressure steam at about 400° F. Surrounding that nickel pipe a copper pipe contained the liquid hex. The critical spacing between the two pipes where the hex flowed measured only about one-tenth of an inch. Surrounding both pipes a 4-inch pipe of galvanized iron carried water at about 130°, just above hex's melting point, to cool the hex.
Pumps that circulated the water were the only moving parts. “The apparatus was run continuously with no shut down or break down what so ever,” Abelson reported to the Navy early in 1943. “Indeed, so constant were the various temperatures and operating characteristics that practically no attention was required to insure successful operation. Many days passed in which operating personnel did not touch any control device.” To stop the flow of the hex out of a column Abelson simply dipped the bend of a U-shaped metal drain tube into a bucket of dry ice and alcohol, which froze the hex and plugged the tube. A flame to warm the tube started the flow again.
Abelson's January 4, 1943, report, submitted jointly with his NRL colleague Ross Gunn, indicated that uranium could be enriched within a single thermal-diffusion column from its natural U235 content of 0.7 percent up to 1 percent or better. With several thousand columns connected in series Abelson thought he could produce 90 percent pure U235 at the rate of 1 kilogram per day at a total construction cost of no more than $26 million. Ninety percent purity was entirely sufficient to make a bomb. (That estimate proved optimistic, however, and equilibrium time for such a cascade appeared to be as long as 600 days.)
Another choice, more in keeping with the Navy's interest in submarine propulsion, emphasized quantity enrichment rather than quality. Abelson proposed building a plant of 300 48-foot columns operating in parallel to make large amounts of slightly enriched uranium immediately. Chicago could use such slightly enriched uranium to advance its pile work, Abelson thought. He did not yet know that CP-1 had gone critical just one month before his report. “Information concerning the many experiments performed by [the Chicago] workers in the last six months has been denied to us,” he complained. “It is vitally necessary that there be an exchange of technical information if proper plans are to be made for future plants.” The NRL had been the first research center Groves visited when he took charge of the Manhattan Project in September 1942. Months before then, Franklin Roosevelt had specifically instructed Vannevar Bush to exclude the Navy from atomic bomb development. Groves followed the NRL's research and Bush encouraged its funding through the Military Policy Committee. But by 1943 the official flow of information on nuclear energy research ran from the Navy to the Army one-way only.
Unofficially, however, several of Groves' compartments leaked. In November 1943 the Navy authorized Abelson to build his 300-column plant. He had searched for a sufficient source of steam — thermal diffusion used volcanic magnitudes of steam, one reason the Manhattan Project had chosen not to pursue it — and located the Naval Boiler and Turbine Laboratory at the Philadelphia Navy Yard. “They were testing good-sized boilers that would go into ships,” Abelson says. “They had the capability of making quantities of steam at a thousand pounds per square inch and they had Navy people standing twenty-four-hour watches to deliver the steam.” The boiler laboratory's waste steam would supply his 300-column plant, but before scaling up that far he planned to test his design by building and operating only the first 100 columns. Construction began in January 1944, with completion scheduled for July. By now Abelson knew more about the Manhattan Project. He knew that the barriers which Houdaille-Hershey had been stripped and reequipped to manufacture were not yet passing inspection and that K-25, the gaseous-diffusion plant, was therefore woefully behind schedule. He knew Los Alamos had been founded with Robert Oppenheimer as its director. He knew Berkeley was struggling to make its calutrons work. He saw that his thermal-diffusion process might come to the bomb project's rescue and he was generous enough and worried enough about the war to offer it despite the Army's several previous rebuffs.
He chose not to work through the limited official channels that the Army and the OSRD had devised to constrict the flow of information. “I wanted to let Oppenheimer know what we were doing. Someone in the Bureau of Ships knew one of the people in the [Navy] Bureau of Ordnance who was going out to Los Alamos. I remember that I met the man at the old Warner Theater here in Washington, up in the balcony — real cloak and dagger stuff.” Abelson briefed the BuOrd officer about the plant he was building. He said that he expected to be producing 5 grams a day of material enriched to 5 percent U235 by July. This vital information the BuOrd man carried to Los Alamos and passed along to Edward Teller. Teller in turn briefed Oppenheimer. Oppenheimer apparently conspired then with Deke Parsons, the Hill's ranking Navy man, to concoct a cover story: that Parsons had learned of the Abelson work on a visit to the Philadelphia Navy Yard. With the Navy thus protected, Oppenheimer on April 28 alerted Groves.
Oppenheimer had seen Abelson's January 1943 report only a few months previously, a year after it was written. He was not impressed. Like his colleagues Oppenheimer had considered only those processes that enriched natural uranium all the way up to bomb grade, a requirement thermal diffusion could not efficiently meet. Now he realized that Abelson's process offered a valuable alternative, the alternative Abelson had proposed in his report to help Chicago advance its pile work: slight enrichment of larger quantities. Feeding even slightly enriched material into the Oak Ridge calutrons would greatly increase their efficiency. A thermal-diffusion plant could therefore substitute at least temporarily for the stalled lower stages of the K-25 plant and supplement the output of the Alpha calutrons. Abelson's 100-column plant with the columns operating in parallel, Oppenheimer calculated, should produce about 12 kilograms a day of uranium of 1 percent enrichment.
“Dr. Oppenheimer… suddenly told me that we had [made] a terrible scientific blunder,” Groves testified after the war. “I think he was right. It is one of the things that I regret the most in the whole course of the operation. We had failed to consider [thermal diffusion] as a portion of the process as a whole.” From the beginning the leaders of the Manhattan Project had thought of the several enrichment and separation processes as competing horses in a race. That had blinded them to the possibility of harnessing the processes together. Groves had partly opened his eyes when barrier troubles delayed K-25; then he had decided to cancel the upper stages of the K-25 cascade and feed the lower-stage product to the Beta calutrons for final enrichment. So he was prepared to understand immediately Oppenheimer's similar point about the value of a thermal-diffusion plant: “I at once decided that the idea was well worth investigating.”
Groves appointed a committee of men thoroughly experienced by now in Manhattan Project troubleshooting: W. K. Lewis, Eger Murphree and Richard Tolman. They visited the Philadelphia Navy Yard on June 1 and turned in their conclusions on June 3. They thought Oppenheimer's estimate of 12 kilograms a day of 1 percent U235 optimistic but emphasized the possibility — with 300 columns instead of 100 — of producing 30 kilograms per day of 0.95 percent U235.
Groves thought bigger than that. He had a power plant with 238,000 kilowatts rated capacity coming on line within weeks in the K-25 area at Oak Ridge that K-25 would not be ready to draw on until the end of the year. It was designed to generate electricity to run the barrier diffusers but it made electricity by making steam. The steam could serve a thermal-diffusion plant that would enrich uranium for the Alpha and Beta calutrons until such a time as K-25 needed electricity. Then the permanent K-25 installation could be phased in gradually and the temporary thermal-diffusion plant phased out.
The proposal cleared the Military Policy Committee on June 12, 1944. On June 18 Groves contracted with the
