Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 364, 282-292 (2000)

Previous Section Next Section Title Page Table of Contents

1. Introduction

The Infrared Space Observatory (ISO) has opened the mid- and far-infrared range for high-resolution spectroscopy and provided striking evidence for the presence of crystalline Mg-rich silicates in comets (Crovisier et al. 1997), circumstellar regions (envelopes/disks) around young stars (Waelkens et al. 1998; Malfait et al. 1999) and evolved stars (Waters et al. 1996; Molster et al. 1998, 1999a). In contrast to chaotic silicates (Nuth 1996) and silicate glasses (Jäger et al. 1994; Dorschner et al. 1995; Mutschke et al. 1998), crystalline silicates show a lot of diagnostic bands due to metal-oxygen vibrations which can be used to identify the minerals of cosmic silicates (Jäger et al. 1998; Koike et al. 1993).

In terrestrial environments, crystalline silicates are mainly formed from cooling melts, from aqueous solutions or by hydrothermal alteration of anhydrous silicates. In astrophysical environments, as listed below, crystalline silicates evolve from amorphous low-pressure condensates by the process of annealing that is characterized by the formation of an ordered arrangement of the silicate tetrahedra by thermal atomic diffusion. In the tetrahedral units themselves, only minor changes occur (Thompson et al. 1996). The crystalline phases developing from amorphous magnesium silicates and from silica are given explicitly (see Table 1).


Table 1. Developing crystalline phases and some properties

The annealing process may well happen in the envelopes around AGB stars (Gail & Sedlmayr 1998; Sogawa & Kozasa 1999; Gail & Henning 1999) and therefore deserves great attention in astrophysical context.

During annealing, thermal diffusion finally leads to a rearrangement of the structural units producing long-range order. Diffusion in solids is based on lattice defects that are thermally activated. That means that thermal diffusion and crystallization can also be stimulated by irradiation and OH enrichment of the silicates. The formation of defects can be non-thermally triggered by particle irradiation (`enhanced diffusion', Frank et al. 1979). OH anions incorporated in silicates reduce the viscosity and promote crystallization and phase separation (Scholze 1988). In low-temperature environments such as molecular clouds, a long-duration exposure to cosmic rays may lead to an accumulation of defects. Thus, the structure of such silicate particles might be characterized by a predominantly non-thermal defect concentration. If a cloud core collapses to form a new star, the dust particles are subjected to temperature rises. The non-thermal defect concentration inherited from the irradiation prehistory might enhance diffusion and promote crystallization. Laboratory experiments have shown that amorphous SiO can crystallize under 80 keV He+ bombardment (Walters et al. 1988). The material was heated up to 1120 K during ion bombardment. At this temperature, the non-irradiated samples did not show crystallization.

In cold dust environments (the dust shell around the young Fe star HD 142527 and comet Hale Bopp) partially crystalline silicates have been recently observed that may have been formed by crystallization at low temperature. In the spectrum of HD 142527, indications have been found that cold crystalline silicates are associated with hydrous silicates (Malfait et al. 1999; Wooden et al. 1999; Molster et al. 1999b). Laboratory experiments confirmed that Mg-SiO smokes are transformed into protophyllosilicates by low-temperature annealing in the presence of H2O. This way, Rietmeijer (1995) demonstrated that Mg-SiO smoke was transformed to saponite and talc by annealing at 378 K after 7.5 h.

In this paper, the differences in the thermal evolution of nanometre-sized magnesium silicate smokes and silica on the one hand and glassy micrometre-sized and bulk magnesium silicates on the other hand will be outlined.

Glassy silicates have a homogeneous chemical composition compared to smokes. From infrared (IR) spectroscopy and transmission electron microscopy (TEM), kinetic constants of the annealing process have been obtained. Silica has been studied as an appearing component during annealing of magnesium smoke and as pure silica; rather different timescales of crystallization have been observed. Additionally, in the far infrared wavenumber range, an opacity drop due to crystallization has been studied.

During annealing at 1000 K up to 4 h, Rietmeijer et al. (1986) found that enstatite crystals had evolved from previously formed tridymite and forsterite nanocrystals. In the study by Hallenbeck et al. (1998), enstatite has not been observed after annealing times up to 3 h at 1200 K. These results have to be compared with the evolution of glassy silicates studied in this paper.

Using IR spectroscopy, Hallenbeck et al. (1998) monitored a "stall" phase, in which the spectra remained almost unaltered. But it remained unclear, why this stall phase occurred and if a stall would occur during annealing of glassy magnesium silicates (Rietmeijer et al. 1986; Hallenbeck et al. 1998). We will deal with this issue.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 2000

Online publication: December 15, 2000