ABSTRACT
Evolution of the intermediate Al-Si spinel phase formed while heating of a variety of mullite precursors has been investigated for identification and quantitative estimation. In the mullite-gel-to-mullite reaction series, major future problems related to the mechanism of mullitization and its different paths of formation are due to mullite’s improper identification and characterization. Spinel formation during heating at/around the first exothermic peak temperature was observed in different cases of mullite precursors. Sometimes, the precursor was found to crystallize as a major weakly crystalline phase, and in other cases it occurred as a minor phase along with weakly crystalline mullite phase in a major quantity. The colloidal gel synthesized by Yoldas was probably amorphous and did not crystallizes into pseudoboehmite in the dried gel state. The resulting gel containing boehmite and silica gel first decomposes into amorphous phases, followed by an endotherm in DTA and loss in weight in thermogravimetric analysis.
