Crystallographic Analysis of Trapiche Ruby using Low-Vacuum Field Emission Scanning Electron Microscopy, Electron Backscatter Diffraction Pattern, Raman Spectroscopy, and Field Emission Electron Probe Microscopy

This is the authors' abstract. We don't add key points for this paper.

Trapiche ruby is a ruby showing textures with six arms and six growth sectors. It has been attracting among jewelry field because of its unique appearance after the early works about it [1-3]. Sunagawa et al. reported important scientific work on it. They showed elemental mapping and optical micrographs, and they also discussed the crystal growth mechanism of trapiche ruby [4]. Recently, we reported elemental and crystallographic study of trapiche ruby using modern updated instruments such as micro X-ray fluorescence spectroscope, low vacuum type field emission scanning electron microscope and high reciprocal space resolution X-ray diffraction [5]. Our main result, especially the elemental analysis, had good agreement with the study of Sunagawa and colleagues, but there were still uncertainties about the crystal growth mechanism of trapiche ruby. In this study, we carried out experimental analysis using low-vacuum scanning electron microscopy (SEM), energy dispersive X-ray spectroscopy (EDX), electron backscatter diffraction pattern (EBSD), Raman spectroscopy, and field emission electron probe microscopy (EPMA) to elucidate the microscopic structure, because it may be useful for future discussion about the crystal growth of trapiche ruby. The specimens were mined in the Mogok area of Myanmar. First, the specimens were cut and polished. Then, a field emission type SEM (JEOL IT-700HR) equipped with EDX (JEOL) and EBSD (TSL Velocity Pro) was used in this study. The SEM, EDX and EBSD observations were carried out under the condition of low-vacuum mode with the pressure of 50 Pa; therefore, we could observe the specimens without the deposition of conductors. This procedure was quite useful for collecting EBSD data because the EBSD signals were not scattered by the deposition layer. Laser Raman spectroscopy/microscopy (Renishaw inVia) was used to analyze whether or not there were any compounds other than ruby (corundum) and to collect mapping data of state analysis. FE-EPMA (JEOL JXA-iHP200F) was used to obtain precise elemental mapping of impurities after carbon deposition. Figure 1 shows the results of wide area elemental mapping by EDX. Not only the main elements of ruby (Al, O, and Cr), but also many kinds of impurities such as S, K, Na, Fe, Ti, Ma, Ca, and Si were found. Cr and most of the impurities seem to be segregated in the "arm" region. This result is in agreement with the previous works [4, 5]. A new finding is that Mg and Ca existed in almost identical areas. We think there are two reasons that we could detect these small Ma and Ca: (1) Low-vacuum type SEM enabled us not to use deposition, therefore the X-ray sensitivity was quite good in this study. (2) Modern field emission type SEM has a good spatial resolution, although it is not the high-end one. This part may be a dolomite, whose chemical formula is CaMg(CO3)2. The Raman spectroscopy results mentioned below also support this conclusion. Figure 2 shows the EBSD inverse pole figure (IPF) map of the trapiche ruby. This is the result of fitting the specimen to have a corundum structure. This result shows that the growth sectors sandwiched by the so-called "arms" and the core part have the same crystallographic orientations. They appear like a single crystal ruby. On the other hand, this IPF map shows that the "arm" part is not a single crystal. Basically, most of the Kikuchi patterns in the arm area could not be indexed as corundum, and some of them were indexed as polycrystalline corundum, but the reliability was low. Therefore, it is clearly considered that the arm part is not a single crystal ruby, and it may be an amorphous phase, although we don't deny the probability that it contains some very small polycrystalline particles. Laser Raman micrography was also performed, and the spectrum from the arm area showed no clear peaks (not shown in this paper). Therefore, it is considered that the arm part may be an amorphous phase. Figure 3 is an example of the Raman spectra collected from the trapiche ruby. In this figure, it is clearly seen both corundum peaks and some other peaks. We also collected many Raman spectra from all over the sample and there are some peaks other than corundum such as diaspore, quartz, dolomite, bukovskyite (not shown in this paper). It is believed that this natural Trapiche ruby is not uniform. Figure 4 (a) is a special Raman mapping, not like a simple intensity mapping of some peaks. It is a mapping of the ratio of peak 4 to peak 3 shown in Figure 3. It shows zonal structure. The zonal structure can be clearly seen by taking the ratio map. Figure 4 (b) is the EPMA intensity map of Ti Kα. The average composition of Ti in this specimen was measured by EDX and EPMA, and it is assumed to be less than 0.2 atomic percent. Therefore, it can be concluded that the zone structure is the result of a very precise compositional fluctuation of Ti. We cannot exclude the possibility of fluctuations of other impurities, but we have not detected them, at least at this stage. It seems that each growth sector has each zone structure. This means that each growth sector has grown independently. Based on these new experimental data, we found the following important information: The arm region of trapiche ruby may be an amorphous phase. At least, it is not a single crystalline corundum. The arm region of trapiche ruby contains many kinds of impurities, and its composition is higher than other parts. Our special Raman mapping data and EPMA mapping of Ti show the zone structure in the growth sectors between the arms. It is assumed that each growth sector has grown independently. It is assumed that the arm, which is not a single crystal, is not the seed of the growth sectors between them. Some compounds such as diaspore, quartz, dolomite, bukovskyite may be included in the trapiche ruby. Optical micrograph of the trapiche ruby (a), and elemental mapping data obtained by EDX (b) ∼ (l), respectively. A yellow rectangle in (a) indicates a mapping area. EBSD inverse pole figure (IPF) map of the trapiche ruby. This image is a combined image of eight individual IPF maps. Kikuchi patterns were analyzed as corundum crystals. The crystallographic orientations perpendicular to the specimen surface are shown in this image. One example of the Raman spectra collected from one area of the trapiche ruby. There exist some peaks different from corundum. (The data from RRUFF ID R040096 is used to draw the blue profile in this figure [6].) Zonal structure observed by Raman mapping (a) and EPMA mapping (b), respectively. Red rectangles represent the same area of the specimen. The Raman mapping results in Figure 4 (a) is the result of the intensity ratio of peak 4 / peak 3 in Figure 3. Figure 4 (b) is the intensity mapping of Ti Ka line.