Ash cenospheres in fly ashes from five Australian power stations have been characterized. The experimental data show that ash cenosphere yield varies across the power stations. Ash partitioning occurred in the process of ash cenosphere formation during combustion. Contradictory to conclusions from the literature, iron does not seem to be essential to ash cenosphere formation in the cases examined in the present work. Further investigation was also undertaken on a series of size-fractioned ash cenosphere samples from Tarong power station. It is found that ∼70 wt% of ash cenospheres in the bulk sample have sizes between 45 and 150 µm. There are two different ash cenosphere structures, that is, single-ring structure and network structure. The percentage of ash cenospheres of a network structure increases with increasing ash cenosphere size. Small ash cenospheres (in the size fractions <150 µm) have a high SiO 2 /Al 2 O 3 ratio, and the majority of the ash cenospheres are spherical and of a single-ring structure. Large ash cenosphere particles (in the size fractions of 150-250 µm and >250 µm) have a low SiO 2 /Al 2 O 3 ratio, and a high proportion of the ash cenospheres are nonspherical and of a network structure. A novel quantitative technique has been developed to measure the diameter and wall thickness of ash cenospheres on a particle-to-particle basis. A monolayer of size-fractioned ash cenospheres was dispersed on a pellet, which was then polished carefully before being examined using a scanning electron microscope and image analysis. The ash cenosphere wall thickness broadly increases with increasing ash cenosphere size. The ratios between wall thickness and diameter of ash cenospheres are limited between an upper bound of ∼10.5% and a lower bound of ∼2.5%, irrespective of the ash cenosphere size.

No abstract