Jf. Banfield et al., TIO2 ACCESSORY MINERALS - COARSENING, AND TRANSFORMATION KINETICS IN PURE AND DOPED SYNTHETIC NANOCRYSTALLINE MATERIALS, Chemical geology, 110(1-3), 1993, pp. 211-231
In this paper we review the distribution, composition and relationship
s among the four TiO2 minerals [anatase, brookite, rutile and TiO2 (B)
] and discuss the possible importance of a fifth phase, TiO2 II, which
has a stability field at high pressure. We also consider the geologic
al application of kinetic data for the anatase to rutile transformatio
n. Previous experimental studies of this reaction were carried out bet
ween 610 degrees and 1000 degrees C. Extrapolation using reported rate
laws over geologic time results in predicted and natural distribution
patterns that are broadly consistent. Experimental data reported here
show enhanced anatase-rutile transformation rates if the anatase is v
ery finely crystalline. We attribute the rapid transformation in these
materials to a lower activation energy (similar to 60 kcal mol(-1) co
mpared to 80-148 kcal mol(-1) reported previously). Rapid coarsening a
ccompanies early, rapid transformation. Declining coarsening rates cor
relate strongly with declining transformation rates and extrapolate to
those calculated from published data for coarsely crystalline materia
l. The first-formed crystals of rutile have average volumes approximat
ely eight times greater than those of coexisting anatase, suggesting t
hat anatase must achieve a critical particle size before it transforms
to rutile. Dopants can affect both coarsening and transformation rate
s. Y ( 1%) retards both coarsening of anatase and its transformation t
o rutile. However, Cr and Ta have very little effect on coarsening rat
es but significantly reduce transformation rates. Cr in anatase lowers
both the transformation activation energy (similar to 47 kcal mol(-1)
) and the frequency factor (similar to 10(11) hr(-1)). Together these
result in comparable transformation rates at temperatures similar to 5
0 degrees C higher than in pure anatase; Ta exerts a similar effect. T
he distribution of the impurities (within particles vs. on the surface
) may largely explain differences between Y and Ta or Cr. Results sugg
est that quantification of the dependence of the transformation kineti
cs on impurities may allow the interpretation of anatase-rutile distri
bution patterns to provide time-temperature information in low-grade m
etasediments.