RADIUM

DONALD P. AMES, FLUOTECH

In 1942, Gen. Leslie R. Groves purchased Belgian Congo pitchblende for the uranium required by the newly formed Manhattan Project. This ore was owned by the Belgium-owned U. S. Radium Corp. When refined, the ore was the radium source for hospital - and - sources. Radon, the daughter generated by radium -disintegration, was condensed in capillaries for personal irradiation sources. Uranium for the Manhattan Project, as U₃O₈, was purified by diethyl ether extraction by Mallinckrodt Chemical, St. Louis, and the radium was returned to U. S. Radium. The company wanted to ascertain that it got all the radium back.

The aqueous solution analyses following radium separation were assigned to the Chemistry Section C-I of the Metallurgical Laboratories at the University of Chicago in June 1945. A control analysis group containing four of the nine members of the Special Engineering Detachment, U.S. Army Corps of Engineers, was formed in July 1945 to develop analyses for aqueous solutions suspected of containing radium. (For more on this project, see Seaborg, Katz, and Manning, editors, "Plutonium Project Record: National Nuclear Energy Series 14B." New York: McGraw-Hill Book Co., 1949.)

	BANG! Colored images of the radioactive emission of a-particles from radium.
	C. POWELL, P. FOWLER & D. PERKINS/SCIENCE PHOTO LIBRARY

Following the successful completion of the radium analyses, it became apparent that the radium half-life had not been determined by direct -counting because of the -particle growth from radon-222 and three polonium isotopes: 210, 214, and 218. The National Bureau of Standards (now the National Institute of Standards & Technology), the repository for radioactivity data, reported 1,590 years for the half-life from an average of several indirect measurements. From an exhaustive literature search, I found 23 indirect measurements of this half-life. I was assigned the measurement of the radium-226 half-life by direct counting in November 1945.

By using the Bateman equations plus a Monroe calculator, I determined the -growth curve for radium based on known half-life values. -Particle growth curves on separate samples were counted for approximately five radon half-lives to determine whether radon diffusion and/or nuclide recoil after -emission was important; the nuclide recoil was found to be important. By using a constant diffusion rate plus variable recoil rates, modified growth curves were constructed. By using these curves with the -growth count for eight hours, the count rate of a known weight of radium after all daughters had been removed could be determined via extrapolation. Then the radium half-life could be determined from this count rate for a known radium weight.

Accurate radium half-life measurements required that I develop micro techniques before proceeding with milligram weights. These techniques included purifying radium by fractional crystallization from aqueous hydrochloric acid, drying purified RaCl₂, quartz fiber microbalance weighing, and preparing deposits for counting, as well as the procedure and counters for eight-hour -counting and extrapolation to zero time. Using arc-spectrograph analysis, the barium content was determined to be less than 0.02%. Then milligram weights of purified RaCl₂ were used to prepare solutions for the half-life determination. Fifty-four different RaSO₄ deposits from the microgram and milligram RaCl₂ solutions were prepared to obtain a half-life of 1,622 years, with an error estimate of 13 years. I completed this measurement in September 1946.

Chemical & Engineering News
Copyright © 2003 American Chemical Society