Neutron thermodiffractometry and synchrotron energy-dispersive diffraction studies of zirconium hydroxide calcination

Abstract

Neutron [l-31 and X-ray [4,5] techniques have been used to study the formation of zirconia from zirconium hydroxide. Synchrotron radiation can now, using dynamic energy-dispersive diffraction (DEDD), shed light on the kinetics of both the hydroxide-to-oxide crystallization and tetragonal-to- monoclinic conversion. Furthermore, different rates of crystallization and conversion have been linked [5] to the preparative conditions of the hydroxide; in particular the final pH of the solution coexisting with the precipitate of zirconium hydroxide. The results here represent an important extension of our dynamic studies. The special advantage of neutron thermodiffractometr!~ (NTD) [6] is that the high scattering/diffraction background associated with incoherent scattering by hydrogen is an effec- tive monitor of the chemical composition of the hydroxide; the scattering background reflects the amount of hydrogen-containing species in the sam- ple. Zirconium hydroxide can be considered as an intermediate between a hydroxide and a hydrated oxide [7]. One possible model [H] depicts the hydroxide as an open polymer in which individual units are linked by double h!.droxy bridges (Fig. 1). Clearly, if such a model were essentially correct, NTD would be expected to monitor the combined loss of H,O and OH during calcination. Zirconium hydroxides were prepared under care- fully controlled conditions [9] at pH 8.37 and 10.35, since these preparations display extremes of crystal- lization behaviour [5]. Aqueous ammonia solution was added to an aqueous suspension of zirconium basic sulphate (Magnesium Elektron Ltd, Man- chester, UK) to the desired pH. the suspension was filtered and washed with distilled water. The sam- ples were dried at 130 "C to produce zirconium hydroxide. Thermal analysis was carried out in both samples at a heating rate of 2 "C min-' in static air. Synchro- tron experiments were undertaken at Daresbury Laboratory, station 9.7, using pellets of thickness approximately 0.5 mm. The pellets were heated at a rate of 20 "C min-' to 400 "C, and the ramp then reduced to 5