PHASE TRANSFORMATIONS IN ALUMINA. Technical Report, May to December 1961

Abstract

e transformation studies showed that synthetic aluminas produced in the laboratory were not transformed to alpha alumina at 400 to 700' deg n the presence of nitric acid and water vapor. Amorphous aluminn produced in the pilot plant was transformed to alpha alumina. This indicated that the amorphous alumina produced in the pilot plant is structurally predisposed to form alpha alumina. Most additives did not appear to have any special effects. Lithium formed zeta alumina, LiAl/sub 5/O/sub 8/. Phosphate formed an unidentified phase. Zinc formed zinc aluminate. Sulfate favored higher water and lower nitrate contents than any of the other additives and formed natroalunite in one sample. The synthetic aluminas were heated in the micro reactor at pressures of 150 to 1300 psi. Most of the synthetic aluminas containing sodium formed alpha alumina while those without sodium did not, which agreed with the pilot plant results. Additives which allowed alpha alumina to form were calcium, iron, and lithium. Potassium, silicate, sulfate and zinc allowed alpha alumina to form when sodium was also present. Phosphate, boric acid, and magnesium seemed to prevent the formation of alpha alumina. Fission products, at 10 times the expected amount, also appeared to prevent formation ofmore » alpha alumina. Phosphate and boric acid deserve additional research to determine concentration limits. New phases were found in the course of the research. A-phase, formed at 150 psi at 400 deg C, was also found to be formed at 1200 psi from aluminum metal and water. A C-phase was found which may be a dibasic aluminum nitrate (Diban), which was different from the crystalline Diban compounds found at Idaho Falls. A B-phase was also found, but not in sufficient purity to characterize. These new phases may be important in the transformation of aluminum nitrate to alpha alumina. Electron diffraction examination was made on particles from selected pilot plant samples. Wedge-shaped sections were examined along a diameter to determine where alpha alumina was initially formed. The random distribution of the most intense patterns indicated that alpha alumina formed over fairly large sections of the particles, (not necessarily at the center or edge) and not randomly in small spots distributed throughout the particle. The electron dlffraction also showed that amorphous alumina had sufficiently small crystallites to appear amorphous to electron diffraction as well as x-ray diffraction. Differential thermal analysis of aluminum nitrate helped prove the complexity of the decomposition. Aluminas from the pilot plant and the synthetic aluminas had different curves. Sample IVA, a pilot plant alumina, had an entirely different curve in the presence of nitrogen dioxide than in the presence of nitric acid vapor, water vapor, or air. This indicated that nitrogen dioxide may have a considerable effect on the decomposition and on the transformation process. Infrared spectra of various samples showed that the nitrate group had ionic rather than covalent bonding. The infrared spectra also indicated that most of the water in the samples was free water; conclusions about the Al-O bond could be obtained with further study. Structural considerations suggest that amorphous alumina produced in the pilot plant might be structurally related to alpha alumina. Radial distribution analysis of pilot plant samples and synthetic aluminas was studied to determine the structure, but the results were inconclusive. Various factors which contribute to formation of alpha alumina are: sodium acting as mineralizer or included in the structure, nitrate and water vapor in intermediate compounds or aiding the mobility of the aluminum and oxygen atoms, and conditions unique to fluidization (e.g., cycling of temperature and cycling of composition in various zones and rate of decomposition). Theoretical considerations were applied to the effect of additives on the type of crystal, on the size of crystallites, and on amor« less