Structural investigation of (130) twins and rutile precipitates in chrysoberyl crystals from Rio das Pratinhas in Bahia (Brazil)

Abstract

We studied V-shaped twins of chrysoberyl (BeAl 2 O 4 ) from Rio das Pratinhas pegmatites near Arataca in the Bahia state of Brazil. The local structure of the twin boundaries was determined using powder X-ray diffraction analysis (XRD), transmission electron microscopy (TEM) methods, and density functional theory (DFT) calculations. To provide the most reliable model for DFT and HRTEM simulations the structure of chrysoberyl was first refined in the orthorhombic space group 62 (Pmnb) with unit-cell parameters: a = 5.4825(1) Å, b = 9.4163(2) Å, and c = 4.4308(1) Å, with 0.5 at% of Fe 3+ present on the Al(2) sites, suggesting an average composition of BeAl 1.99 Fe 0.01 O 4 . TEM study of V-shaped twins showed that the twin boundary lies in the (130) planes, and the angle measured between the crystal domains related by mirror twin operation is ∼59.5°. Rigid structural model of (130) twin boundary in chrysoberyl was refined by DFT calculations, using a pseudo-potential method. The twin boundaries show local enrichment with Ti. Bulk chrysoberyl contains numerous nanosized TiO 2 precipitates with a distorted rutile structure, following the orientation relationship of [001] Ch {120} Ch ||[010] R {103} R . The increase of Ti at the twin boundaries and the formation of rutile-type TiO 2 precipitates in bulk chrysoberyl suggest a transient Ti-exsolution that took place after the twin formation.

Characterization of Oriented Inclusions in Cat's-eye, Star and Other Chrysoberyls

“…More atypically, in one area of the Kerala thin section cut perpendicular to the a-axis, structures appearing as short lines were observed parallel to the <012> directions of the chrysoberyl host, and were often connected to form L-shaped or zigzag patterns (Figure 20). Such an orientation was recently described in a chrysoberyl from Brazil (Drev et al, 2015). In the Kerala sample, these inclusions also were identified as rutile by Raman spectroscopy.…”

“…(For an explanation of the nomenclature referring to crystallographic directions and their respective axes in chrysoberyl as employed in this article, see Box A.) More recently, two series of oriented inclusions described as rutile precipitates along k and -k {021} faces were studied by electron diffraction in a chrysoberyl from Brazil (Drev et al, 2015). The equivalent directions of those elongated structures would be <012>, and the maximum length of the precipitates was 0.05 µm.…”

“…The equivalent directions of those elongated structures would be <012>, and the maximum length of the precipitates was 0.05 µm. The relationship between the crystal structures of the chrysoberyl host and those of the rutile exsolutions was discussed in detail by Drev et al (2015), but the dimensions of the precipitates in the direction parallel to the a-axis [100] has not yet been determined (Prof. A. Rečnik [Jožef Stefan Institute, Ljubljana, Slovenia], pers. comm., 2016).…”

“…As such, the linear structures along the a-axis [100] and at right angles to the L-shaped and zigzig patterns represented the junctions between the different parts of the larger platy rutile inclusions. Moreover, taking into account the description by Drev et al (2015) of rectangular structures along <012> as intergrown rutile clusters, it could further be concluded that: (1) the larger platy inclusions or complex platelets described above were rutile clusters, and (2) the linear structures elongated parallel to the a-axis [100] represented the edges or interconnections between the different inclined parts of the platy clusters. In other words, the platy rutile clusters were composed of narrow inclined sections elongated parallel to the a-axis [100] that caused the platelets to appear 'striped' when viewed perpendicular to the a-axis.…”

Rutile inclusions in garnet from a dissolution‐reprecipitation mechanism

“…The observed COR(s) between rutile and the host crystal is then determined by the alignment of the close‐packed oxygen arrays of the two structures. Orientation relationships controlled by the oxygen‐packing scheme are also noted in oxide‐silicate systems, such as topotactic relationships of ilmenite, hematite or magnetite in olivine (Champness, ; Dobrzhinetskaya, Green, & Wang, ; Hwang, Shen, Chu, & Yui, ; Hwang et al, ; Risold, Trommsdorff, & Grobéty, ), magnetite in clinopyroxene (Hwang et al, ), rutile in biotite (Shau, Yang, & Peacor, ), as well as rutile in chrysoberyl (Drev, Komelj, Mazaj, Daneu, & Rečnik, ).…”