Development and evaluation of alumina calcination

  • Ian John Bennett

Student thesis: Doctoral ThesisPhD


This thesis focuses on a number of aspects governing the transformation of gibbsite, via intermediate phases, to α-alumina. These aspects include the size and morphology of the gibbsite grains, the influence of additions of foreign elements, the effect of a mechanical treatment of the gibbsite prior to calcination, and combinations of these factors. The materials were characterised by scanning electron microscopy, X-ray diffraction and surface area measurements. For some of the calcined materials an attempt was made to sinter the powders to a dense body to investigate if any of the treatments during calcination had an effect on this process.

The literature review covers the current state of understanding of the production of bulk alumina powder by the Bayer process and the phase changes seen on calcination of precursors to the stable α-alumina phase. A detailed description of the phase changes is given and the various routes and conditions necessary for the transformations to occur are considered. The transformations are examined in relation to the morphology of the crystals and the variables controlling the phase transformation route are discussed.

Calcination in air showed that the size of the gibbsite grain governs the calcination route taken to reach α-alumina. The standard gibbsites used in this work show a mixed calcination sequence transforming both via the boehmite phase, followed by the γ, δ and θ phases, and via the χ and κ phases. The formation of boehmite is attributed to retention of water vapour within the grain.

Differences in morphology of the starting materials showed that for the range of materials seen, the morphology of the grain is less important than its size. The super fine material confirmed that a small grain size transforms via the non-boehmite route only, with the other gibbsites taking intermediate routes as for the standard gibbsites.

Of the additions made prior to calcination, aluminium fluoride was found to reduce the transformation temperature to α-alumina by approximately 300°C. Other additions had little effect on the transformation temperature although a reduction in grain size was seen with aluminium chloride. It was found that good mixing of the alumina fluoride was essential to obtain reliable and reproducible results. This is due to the small amounts of additive that are needed and the sensitivity of the process to concentration variations. Mineralisation of a range of gibbsites showed that the presence of sodium in the starting material was crucial in reducing the calcination temperature. This led to the conclusion that the sodium and fluoride react to form a liquid phase. The presence of a liquid phase increases the mobility of the aluminium and oxygen atoms resulting in a reduction of the transformation temperature. Fluoride additions to the gibbsites with different morphologies showed that the presence of sodium was the governing factor in reduction of the transformation temperature.

Milling of the starting materials showed that there was a small reduction in the transformation temperature between some of the phases. The energy involved in milling leads to activation of the gibbsite. This activation takes the form of a reduction in the grain size and in a reduction of the crystallinity seen in the XRD pattern.

Fluoride additions during the calcination of sapphire with a standard gibbsite powder showed preferential grain growth. It was possible to initiate growth of small plate-like crystals on the polished surface of a piece of sapphire parallel to the basal plane. Crystal growth was also seen in scratches on a polished surface perpendicular to the basal plane.

Date of Award21 Feb 2001
Original languageEnglish
Awarding Institution
  • University of Bath
SupervisorRonald Stevens (Supervisor)

Cite this