Although Facilities Available in Sri Lanka Were Limited, Great Care W.As
Total Page:16
File Type:pdf, Size:1020Kb
CHAPTER 7 - DISCUSSION AND CONCLUSIONS Although facilities available in Sri Lanka were limited, great care w.as taken to fulfill the objectives of the project within the accuracy needed, by performing as many tests as possible in Sri Lanka and elsewhere. The origin of Sri Lankan corundum was described to be a metamorphic process by several previous authors such as (Munasinghe and Dissanayake, 1981, Rupasinghe and Dissanayake, 1987, Dahanayake and Ranasinghe,. 1985, Hapuarachchi, 1989 ) and they have carried out considerable amount of work in this respect. The objective of this study is to explore the validity of these theories further. Several criteria associated with Sri Lankan corundum found in alkali magmatic terrains were observed during this study. During this study certain phenomena was confirmed with analytical data. These are directly related to the determination of origin, hence they are summarized as follows {(i), (ii), (iii), (iv)(v)}. (i) Once the corundum is formed at peak metamorphic conditions and is also subjected to retrograde alteration, minerals such as diaspore, chlorite and mica could be seen surrounding the corundum and within its fractures and fissures (Fernando et al., 2001). But within the calc-silicate rock plagioclase acts as a shield, thus preventing such alterations. Corundum crystal with its outer surfaces completely enclosed by the plagioclase coat was found to be intact, without any damage, in spite of the mineral reactions. While the corundum crystal, which was partially surrounded by re-crystallized plagioclase, had the uncoated portions damaged due to 136 mineral reactions. This process could be described as plagioclase shielding effect on corundum, because in a way it safeguards the corundum crystal from damages caused by mineral reactions. The crust was confirmed as plagioclase by EPMA analysis. This feature had been observed in corundum from alkali magmatic terrains in countries such as South West Rwanda, Thailand, China and Australia (Krzemnicki et al., 1996 and Sutherland et al., 1998). It should be noted that the same feature is now being reported in Sri Lanka thereby associating corundum with magmatic conditions. (ii) Surface features of needle-like pattern often radiating from a central point (resembling a duck's foot) were observed in Sri Lankan corundum. Up to now a similar description had been recorded only in alkali magmatic terrains, as a surface pattern on corundum revealing the remnants of its former coatings, originally crystallized from the coarse basalt host, but have since weathered (Krzemnicki, et. al., 1996). Since the same conditions were observed in Sri Lanka, the magmatic nature could be applied to them as well. (iii) Zircon inclusions were quite abundant in Sri Lankan corundum, yet no record was found describing any of these zircon clusters (aspect ratios for the majority of zircon was in the range of 1.4 to 2.3). Present investigation revealed zircon clusters as an inclusion in corundum. This phenomenon of aspect ratios ranging from 1.2 to 1.8, had been observed earlier, confined to corundum associated with alkali basaltic 137 deposits of Australia and China (Guo et al., 1996). Since the ratios observed in Sri Lanka too were in close proximity to these the magmatic properties of corundum were evident. (iv) Radial cracks surrounding the zircon inclusions were observed in corundum found within both the granulitic gneiss and calc-silicate rocks of Balangoda region. In a way this process was partially responsible for the depletion of its quality. This feature too had been observed earlier, only in relation to corundum located within alkali basalt terrains (Saithai and Rankin, 1999) Now Sri Lanka too witnessed the same feature and its magmatic nature. (v) Chemical fingerprinting was a method developed by Sutherland et al. (1988) to ascertain the origin of corundum (magmatic or metamorphic). This is based on the quantitative EDXRF analysis of elements such as Cr and Ga. (the 'metamorphic' suites have high chromium/low gallium chemistry, with Ga203 contents<0.01 wt% and Cr203/Ga203 ratios above 3. The 'basaltic' suites show higher Ga203 (up to 0.04 wt%) and Cr203/Ga203 ratios below 1}. Out of several samples selected and analyzed, the results of two samples revealed that they were of magmatic category while another two were of intermediate origin between metamorphic and magmatic. Due to the plagioclase shielding effect, the corundum retains its gem quality; lack of it may leave room for cracks and fissures to be developed as a result of mineral reactions, thereby lowering the quality, in par with 138 industrial corundum. Radial cracks due to zircon inclusions too would contribute to decreasing the quality of corundum. In case the zircon inclusions are large and numerous, the chances of obtaining gem quality corundum is rare. These phenomena would facilitate in exploring gem quality corundum. These days the alluvial deposits are dwindling, therefore this could be a deciding factor to find insitu deposits containing gem quality stones. Chemical fmgerprinting is the methodology used in some other countries to locate the origin of corundum collected from different parts of the country. Sri Lanka being a small country, it was not viable to utilize such sophisticated methods. Yet the fact remains that the corundum properties differ from locality to locality. Experienced gem dealers seek gems from certain areas, because they are aware that those stones, once cut and polished would achieve a high standard. This is evident in the case of geuda heat treatment. Trace elements are the key ingredients responsible for this condition. Chemical fingerprinting is a method based on trace element analysis, and it is an ideal method to sort out gems of particular localities. In addition to the above, some spectacular reaction textures were observed in two cases. The reactants and products of these were confirmed by analytical data. In the first observation, formation of spinel from phlogopite mica and corundum was observed and the reaction was a balanced one. In this instance it was a metamorphic reaction and without doubt proved the corundum to be of metamorphic origin. 139 In the second observation corundum, hexagonal shaped spinel and instances of corundum combined with spinel were observed. As there was spinel with hexagonal outline intact the reaction must have been with corundum to form spinel and phlogopite mica. Although the direction of the reaction (corundum <==> spinel) could not be predicted with the available data, the reaction was reliable as far as the analysis was concerned. This too is a metasomatic process, which proved the origin of the corundum in this instance to be metamorphic. Sapphirine as an inclusion was detected within one source rock of corundum. The cation proportions were used to plot oxide molecular sums on the MgO + FeO, A120 + Fe203, Si02 composition diagram. The mol.% oxide sums MgO + FeO = 35.41, Al20 + Fe203 = 48.56, Si02 = 15.81, and consequent mean composition was closer to the 7:9:3 composition [(Mg,Fe)0: A1203: Si02], often quoted in literature (Sengupta et al., 1990). Furthermore, sapphirine was found in contact with spinel and the sapphirine/spinel thermometry was calculated (Owen and Greenough, 1991). Calculations based on the sapphirine-spinel pair gave crystallization ft 0-4- temperature of corundum of Sri Lanka as 761 C (Fe calculation). It seemed that fluid Inclusion study of Sri Lankan corundum had not been carried out by the authors mentioned previously, other than the foreign researchers. During the study it was possible to confirm the previous theories, such as the statement that primary and secondary fluid inclusions of corundum contained more or less pure C02 (Bruder, 1995) and the common daughter minerals found in the Sri Lankan corundum were graphite and diaspore (Schmetzer and Medenbach, 1988), with the help of 140 analytical data. In addition sulphur (418.7, 577.9 and 749.8 cm"1) was detected as a daughter mineral. Necking was also observed and the process was similar to the one described by Grubessi and Marcon, 1986. Mainly ten forms (shapes) of fluid inclusions of the Sri Lankan corundum were identified. Finally a fluid inclusion classification for Sri Lankan corundum was constructed on the basis of form (shape), size and distribution pattern of those detected. According to this classification primary inclusions were divided into three categories (primary type 1, 2, 3) and secondary inclusions into one category (secondary type 1). Presence of pure C02 reveals that the corundum had not been subjected to heat or radiation treatment, because in the latter stage C02 is not present as it is (Koivula, 1988). This principle could be utilized to detect such treatments, because in Sri Lanka the dealers usually do not divulge such treatments. C02 detection in a stone could help to recognize its non-treated status. Present investigation of Sri Lankan gem minerals and rocks could highlight the factors described in (i), (ii), (iii), (iv), (v) and the evidence related to metamorphic growth of corundum. These factors cannot be applied to the entire country, because the investigation did not cover the entire landmass, hence there is always room for both these types to exist. If it were realized that all the factors in both these categories were observed in the context of one single area (Balangoda 141 region), then of course it would make sense. Considering these criteria it should be reasonable to conclude that the origin of corundum of Sri Lanka must be a combination of metamorphic and magmatic. As for the future, fluid inclusion study would be a useful and an essential study, because during the present investigation no fluid other than C02 was encountered and this became a handicap to the project, owing to the total homogenization temperature being equal to that of C02, there by restricting the possibility of finding the pressure temperature time path and the subsequent cooling uplift period of corundum.