The following report covers the construction materials selected for laboratory corrosion studies and the determination of their corrosion rates in process solutions.

Abstract: "X-ray and microscopic examination of slags from uranium metal production bombs which gave good and bad yields in the plant at St. Louis have indicated the significance of uranium oxide formation. Apparently the oxide (UO2) results from reaction of the UF4 with linear material, particularly with moisture in the liner."

Infrared spectra of solid NaXO2(Ac)3, with X=U, Np, Pu, and Am, have been observed. From the symmetric and asymmetric stretching frequencies of the O-X-O groups, approximate X-O force constants have been calculated and were found to decrease in the order kU—O>kNp—O>kPu—O> kAm—O, the respective values being about 0.705, 0.698, 0.675, and 0.612 megadyne/cm. From the cell constants for NaXO2(Ac)3 it is apparent that the X-O bond distance decreases in the same order—RU—O>RNp—O>RPu—O>RAm—O. Thus, a decrease in bond distance appears to be accompanied by a decrease in force constant, probably because the bond, though shortened by contraction of the electron shells of the metal, is weakened by interaction with the extra valence shell electrons.

Chemical and engineering data were obtained for the feed digestion system and the extraction-scrub step of the Monex tributyl phosphate solvent-extraction process for recovering thorium and uranium from nitric acid-digested unclarified monasite sludge. Tests of the recommended conditions in a 2-in.-dia pulsed column demonstrated that thorium losses were approximately 1.2% and uranium losses, 1.5%. The flowsheet is workable but is not necessarily optimum.

This technical report presents the process design for the thorium semi-works plant which will be located at the Feed Materials Production Center, Fernald, Ohio. It contains the process information required for the layout and detailed mechanical design of the thorium plant.

The purpose of this report is to provide calculated liquid flow pressure drop data for the 1706-KER recirculation loops.

Chloride-induced stress-corrosion cracking has been encountered in the Homogeneous Reactor Test during the preliminary testing. The rector is constructed of austenitic stainless steels. It is unique in that it will operate at 250 to 300 C with an aqueous uranyl sulfate solution fuel containing 200 to 500 ppm of dissolved oxygen. The cracking has occurred in a secondary system used for detecting leaks in the flanged joints of the primary systems and in the grooves of flanges in the primary systems. Tubing used in the leak-detection system was found to be contaminated with chloride introduced during manufacture.

A second test of the mockup of the Zircaloy - stainless steel transition joint as used in the HRT reactor vessel has been completed. The joint and bellows have now received 104 thermal cycles and 148 mechanical deflections. The joint and bellows have functioned properly; corrosion damage has been negligible, except for a small area on the bellows which has undergone pitting attack. Long-term runs with uranyl sulfate solutions of the concentration proposed or use in the HRT have shown the solution to be stable at 300 C. Substituting heavy water for normal water caused no difference in either corrosion or solution stability. Experiments in which chromic acid was used to pretreat stainless steel have shown that, under certain conditions, the pretreated film can exist in uranyl sulfate solutions at flow rates in excess of the critical velocity for relatively long periods of time. The practicability of using titanium inserts in high turbulent areas of stainless steel loops to minimize corrosion has been demonstrated. The corrosiveness of beryllium sulfate solutions containing dissolved uranium trioxide has been determined at 250 and 280 C. Laboratory studies with regard to stress-corrosion cracking have shown that high stressed type 347 stainless steel will crack …

Resistance of carbon steel to corrosion in oxygenated high-temperature (250C) water was unexpectedly good at high oxygen concentration. Pertinent literature, critically examined, and toroid experiments indicted that at low oxygen concentration attack did increase with concentration, but as oxygen concentration was sufficiently increased, more protective films were formed on the metal. Some corrosion factors in the application of carbon steel to nuclear reactors systems are discussed.

After several failures had occurred in the HRT leak detector system, several lengths of this tubing were removed for metallurgical examination. The tubing was of type 304 stainless steel and was 1/4" in diameter with a 0.065 wall. The tubing had been purchased as three different lots, the first in 45 ft. lengths and the other two as standards lengths. Tubing from the first lot was used primarily for the shield penetration and, therefore, sections of it are present in all lines of the system. It appears that chloride contamination entered the system in a portion of the first lot of tubing used for the shield penetration. The exact source of the chloride cannot be determined, but after considering the results and visiting the manufacturer's plant, it appears most likely the contamination was during the manufacturing process.