The crystal structure of ammonioleucite was refined by a Rietveld analysis using X-ray diffraction data of the type specimen. The refinement with a space group I4_l/a was converged at R_wp=12.12, R_p=9.39, R_F=4.86, R_B=7.71, and R_e=2.55% with lattice parameters of a=13.2106(6) and c=13.7210(7) Å. Ammonioleucite is isostructural with leucite. The (Si, Al)O₄ tetrahedra form 4-, 6-, and 8-membered rings by sharing their apical oxygen atoms. The crystal structure consists of the three-dimensional framework of the rings of the tetrahedra. Ammonium ions are located in cavities in the framework. The (NH₄)⁺ ions in these cavities are larger than the corresponding K⁺ ions in leucite, causing the cavities to be enlarged relative to those in leucite. Among the 4-, 6-, and 8-membered rings of (Si,Al)O₄ tetrahedra, only the 8-membered rings have diameters large enough to allow the migration of exchangeable K⁺, (NH₄)⁺ and Na⁺ ions.

View full abstract

Evaporation experiments of forsterite single crystals were carried out in H₂ gas (1.4×10⁻⁵ bar H₂) at temperatures ranging form 1200 to 1450°C. Forsterite evaporated congruently. The measured evaporation rates of {010} surfaces were j_Fo=2480_{[mole cm⁻²sec⁻¹]}exp(−372_{[kJ mole⁻¹]}⁄RT). A model on the evaporation rates of forsterite in the system Mg₂SiO₄–H₂, where free evaporation-dominated (RED) and hydrogen reaction-dominated (HRD) regimes are included, was revised based on the Hertz-Knudsen equations with equilibrium vapor pressures. The measured evaporation rates are less than the ideal rates calculated by the present model, showing that kinetics at the evaporating surfaces constrain the evaporation rates. The evaporation coefficients, which are defined as the ratios between measured and calculated rates, were 0.04–0.12. Data on present and previous experiments of forsterite evaporation in H₂ gas and vacuum were compiled, and compared with the calculated rates. The evaporation coefficients varies from about 0.03 to 0.2 in a wide range of temperature and H₂ pressure. No specific relation with temperature or total pressure was recognized, and the evaporation coefficient can be roughly regarded as 0.1 both in the FED and HRD regimes.

View full abstract

The microstructure of fibrous iron-bearing brucite (nemalite, (Mg,Fe)(OH)₂) has been investigated using high-resolution SEM and TEM to reveal the origin of its morphology. The fiber consists of aggregates of flat, lath-like brucite single crystals elongated parallel to <10-10>. Chrysotile tubes, which are also parallel to the fiber, frequently exist, mainly at the grain boundaries of brucite and surrounded by brucite crystals. It is suggested that this unexpected morphology of brucite was formed by crystal growth along the preexisting chrysotile tube surfaces. However, the interface between brucite and chrysotile is not coherent and the arrays of hydroxyl on the surfaces of two minerals are rotated by 30°C to each other. Such orientation relationship is often reported in aggregates of sheet silicates. Brucite crystals frequently contain 180° rotational twins on (0001).

View full abstract

The multi-site regular solution model is applied to the intracrystalline exchange reaction of Fe²⁺ and Mg²⁺ between M1 and M2 sites of olivine. Effects of Ca on the Fe–Mg intracrystalline distribution coefficient K_D[=(X_Mg^M2X_Fe^M1)⁄(X_Fe^M2X_Mg^M1] are derived from this solution model, which indicates that Fe preferentially orders onto the lager M2 site accompanying with increasing Ca. This reaction would be assumed to be temperature-independent. The difference between Fe–Mg mixing parameters of M1 and M2 sites is very small and temperature-independent, so that the fayalite-forsterite olivine would be approximated as a binary regular solution.
The miscibility gaps between high-Ca and low-Ca olivines are sufficiently described by the disordered multi-site regular solution model, which is very similar to a symmetric regular solution model. The present estimates for Margules parameters W_MgCa^M2 is consistent with Davidson and Mukhopadhyay’s (1984) analysis of miscibility gap in Ca–Fe–Mg olivines, and is slightly less non-ideal than the Adams and Bishop’s (1985) results estimated from the mon-ticellite-forsterite solvus. The Ca–Fe mixing is more non-ideal than the previous estimates.

View full abstract