Chapter 4 : World War II: 1939-1946
The Prewar Years
The 1930s had been a period of great changes on the international scene, in the United States, and at MIT. Military confrontations and invasions in the Far East and Europe were warnings of armed conflicts to come, The Fall of 1939 saw the start of World War II.
...
My war work was mainly with the Chicago division of the Manhattan Project. It began at MIT with the development of vacuum furnaces for melting and casting the pyrophoric uranium powder produced by Metal Hydrides. Arthur Compton asked me to come to Chicago as Chief of the Metallurgy Section of the Metallurgical Laboratory. I had four very competent group leaders ... Also I maintained the contact with other metallurgical laboratories at Battelle Institute, National Bureau of Standards, Ames, Iowa and MIT. ... At MIT the work was in part a continuation of my earlier work. When Cyril Smith at Los Alamos complained that he could not buy sound beryllium rods, the MIT group under Morris Cohen and Al Kaufmann modified the uranium furnace to melt and cast the commercial product. This was then extruded by a method developed by Kaufmann of using a copper or soft iron jacket as a lubricant around the beryllium billet. The coating was then stripped off leaving a sound beryllium rod.
A principal problem at Chicago was the protection of fuel elements against corrosion by the cooling water of the Columbia River. After many trials of dipped or plated coatings it became evident that the uranium "slugs," about 8 inches long and 1 1/4 inches diameter, would have to be encased in a metal jacket. Aluminum was the obvious metal. John Howe's studies of corrosion had demonstrated that an aluminum cover and the aluminum tube were adequately resistant to Columbia River water. The aluminum "cans" were impact-extruded by Alcoa. We had to develop a method for canning the slug such that complete thermal contact existed at every point. This problem was solved by Al Greninger by dipping the slug into a molten zinc-base solder and inserting it, hot, into the can. The excess aluminum was trimmed oft the end crimped on an aluminum end piece and the closure completed by the then very new heliarc welding. The success of the Hanford reactors depended upon 100 percent perfection in the pieces. One failure could have meant catastrophe. Imperfection in the bond between slug and can could cause a hot spot with failure of not only the piece but the whole pile. The inspection method known as the frost test was developed by AI Kaufmann at MIT, perfected in Chicago and set up at Hanford. The pieces were immersed over dry ice in acetone. When brought out they frosted in the air. At any spot where the bond was imperfect the frost melted rapidly and any such piece was discarded. After the war was over and at the time the Smythe Report was written there had not been one failure in the Hanford reactors. In my opinion both Greninger and Kaufmann should have had the Medal of Merit; but then no one asked me.
With Hanford in operation my attention turned to helping Los Alamos. I have mentioned Kaufmann's perfection of beryllium. The next was production of crucibles for the vacuum-melting of plutonium and enriched uranium. Several ceramic laboratories declined to undertake it and I persuaded F.H. Norton at MIT to take it on. It was about this time also that I began spending most of my time at MIT with frequent visits to Los Alamos. For melting uranium we perfected a high-purity magnesia crucible and these were made in several sizes and considerable numbers. At the start it was thought that plutonium would have to be oxygen-free. Our experience with other metals indicated that a non-oxide crucible would be needed. Leo Brewer at Berkeley showed that the most stable sulfides were CeS and ThS and he made a number of small crucibles of each. We concentrated on CeS of larger sizes. Cerium sulfide was not available and, following Brewer, we converted Ce~2~O~3~Ce~2~ O~3~, to CeS, by reaction with H,S; this was hydrogen-reduced to Ce,S, and the final step was further reduction with metallic cerium. The resulting brassy product was ground and shaped into crucibles which were fired in vacuo using a graphite susceptor at 17000 C. We had progressed up to about 4 inches high by 2 inches diameter when it was discovered that a little oxygen would do-no harm and thereafter plutonium was melted in our MgO crucibles.
...