30 40 50
Two Theta (degrees)
two theta (degrees)
Figure 1. Left: crystallization of MgSi03, bottom to top: initial 1000K; after 3 hrs; after 19.5 hrs; initial 103 IK; after 24.5 hrs; initial 1043K; after 24 hrs; at 1083K; at 1143K; initial 1173K; after 14.5 hrs. Right: selected data, bottom to top: Initial 1000K; after 19.5hrs; 1143K; after 14.5hrs at 1173K
and is confirmed experimentally . Below -70% removal, the removal curves approximate those of simple diffusion with an effective diffusion coefficient of 1.4xl0"7 cm2 s"1, which is much less than the expected value of 6xl0"6 cm2 s"1 for simple diffusion . Taken in conjunction with deviations of the removal curves from simple diffusion above 70% removal, this implies that H2 outflow is controlled by the reaction rate and not the diffusion coefficient  and is supported by an activation energy of 260-266 kJ mol"1  which matches the SiO-H bond energy of 264 kJ mol"1 , Dehydroxylation is thus controlled by the breaking of this bond. X-ray absorption spectroscopy at the Si K shell absorption edge in annealed Mg silicates  revealed shifts in the energy of the absorption features, which were greatest below the crystallization threshold. Such shifts are characteristic of changes in the oxidation state of the absorbing species (i.e. Si polymerization) , confirming that most alterations to the local atomic environment of Si occur early on. The closeness in threshold temperature for (1) and the initial temperature in our experiment where early alteration to the amorphous component occurred, strongly suggests that dust grains annealed at low-end temperatures will be likely to retain an amorphous component at higher temperatures.
The net effect of (1) is to increase the overall SiC>2 network by reconnecting intertetrahedral Si-O-Si bridging bonds, which raises the average number of bridging oxygen atoms per tetrahedral unit (BO/T). Since forsterite comprises units with BO/T-0, the extent of its development is limited by the need to break bridging bonds on units with BO/T> 1. Amorphous silicates however will contain a distribution of BO/T values ranging 0..4 , so that low-end annealing will initially promote the ordering of BO/T= 0 species. Crystal growth thus stalls when the need for more BO/T-0 units can only be met by breaking bonds for units with BO/T>l. Meanwhile, during the initial stages of annealing, the bulk BO/T for this component will have been increased as a direct result of annealing and its average BO/T raised towards that of Si02 (BO/T=4). The stall thus persists until enough energy has been input to the system to allow (the increased number of) bridging bonds to be broken, whereupon the amorphous phase breaks down and further crystal growth becomes possible. This trend is visible in our structural data and is corroborated by the previous observation that the higher the initial annealing temperature, the shorter the spectral stall period .
Was this article helpful?