K.S. Viljoen

Origin of Ti-rich garnets in the groundmass of Wajrakarur field kimberlites, southern India: insights from EPMA and Raman spectroscopy

Although Ti-rich garnets are commonly encountered in the groundmass of many alkaline igneous rocks, they are comparatively rare in kimberlites. Here we report on the occurrence of Ti-rich garnets in the groundmass of the P-15 and KL-3 kimberlites from the diamondiferous Wajrakarur field in the Eastern Dharwar craton of southern India. These garnets contain considerable Ti (11.7–23.9 wt.% TiO2), Ca (31.3–35.8 wt.% CaO), Fe (6.8–15.5 wt.% FeOT) and Cr (0.04–9.7 wt.% Cr2O3), but have low Al (0.2–5.7 wt.% Al2O3). In the case of the P-15 kimberlite they display a range in compositions from andradite to schorlomite, with a low proportion of grossular (andradite(17.7–49.9)schorlomite(34.6–49.5)-grossular(3.7–22.8)-pyrope(1.9–10.4)). A few grains also contain significant chromium and represent a solid solution between schorlomite and uvarovite. The Ti-rich garnets in the KL-3 kimberlite, in contrast, are mostly schorlomitic (54.9─90.9 mol %) in composition. The Ti-rich garnets in the groundmass of these two kimberlites are intimately associated with chromian spinels, perhaps suggesting that the garnet formed through the replacement of spinel. From the textural evidence, it appears unlikely that the garnets could have originated through secondary alteration, but rather seem to have formed through a process in which early magmatic spinels have reacted with late circulating, residual fluids in the final stages of crystallization of the kimberlite magma. Raman spectroscopy provides evidence for low crystallinity in the spinels which is likely to be a result of their partial transformation into andradite during their reaction with a late-stage magmatic (kimberlitic) fluid. The close chemical association of these Ti-rich garnets in TiO2-FeO-CaO space with those reported from ultramafic lamprophyres (UML) is also consistent with results predicted by experimental studies, and possibly implies a genetic link between kimberlite and UML magmas. The occurrence of Ti-rich garnets of similar composition in the Swartruggens orangeite on the Kaapvaal craton in South Africa, as well as in other kimberlites with an orangeitic affinity (e.g. the P-15 kimberlite on the Eastern Dharwar craton in southern India), is inferred to be a reflection of the high Ca- and high Ti-, and the low Al-nature, of the parent magma (i.e. Group II kimberlites).

The Pipe-15 kimberlite: A new addition to the Wajrakarur cluster of the Wajrakarur kimberlite field, Eastern Dharwar craton, Southern India

This paper serves to report, for the first time, on the chemical composition of primary groundmass minerals occurring in a newly discovered kimberlite located 150m NE of the well known Pipe 2 kimberlite in the Wajrakarur cluster of the Wajrakarur kimberlite field, Eastern Dharwar craton, Southern India. These mineral compositions are also compared with newly acquired analyses of groundmass minerals in the Pipe 2 kimberlite. The kimberlite in this newly discovered intrusion (here referred to as Pipe 15) is characterized by a magmatic, microcrystic rock texture, consisting of microcrysts of serpentinised olivine in a finer groundmass rich in phlogopite, clinopyroxene, spinel, perovskite and apatite (in a base of serpentine). The rock is intensely altered. Phlogopite in the Pipe 15 kimberlite has comparatively low Al and higher Fe than phlogopite in Pipe 2 and (on this basis) is markedly different to phlogopite from Pipe 2. Spinels in Pipe 15 are also compositionally markedly different to that in Pipe 2 in that lower Ti/(Ti+Cr+Al) and Fe/ (Fe+Mg) are seen in spinels from Pipe 15, when compared to spinels from Pipe 2. Furthermore, Fe-poor, primary (groundmass) clinopyroxenes are abundant in Pipe 15, while it is almost totally absent in Pipe 2. In view of these compositional differences, it is here considered likely that Pipe 15 represents a new, separate intrusion and that it is not simply a satellite of Pipe 2, in the Wajrakarur kimberlite cluster. The compositions of the phlogopites, spinels and clinopyroxenes from the Pipe 15 kimberlite, as analyzed in the present study, are strikingly similar to that of the groundmass phases (phlogopite, spinel and clinopyroxene) in Group 2 kimberlites (orangeites) from the Bastar craton in Central India, as well as for typical orangeites from the Kaapvaal craton in South Africa. A genetic relationship to Group 2 kimberlites (orangeites) can therefore be inferred in view of modern mineralogical genetic classification. The recent discovery of Late-Cretaceous orangeite in the nearby Timmasamudram kimberlite cluster is also indicative of Group 2 kimberlite (orangeite) magmatism in this region of kimberlite intrusions, and serves to support the classification proposed here for Pipe 15.

A mineralogical perspective on the recovery of platinum group elements from Merensky Reef and UG2 at the Two Rivers mine on the Eastern limb of the Bushveld Complex in South Africa

Published studies dealing with the process mineralogy of Pt mines on the Bushveld Complex is generally limited to the Western Bushveld. The recognition by mine management that another resource, in addition to the Upper Group 2 (UG2) reef currently being mined at the Two Rivers platinum mine (TRP), is urgently required in order to extend the life of mine, presented an opportunity to conduct such a study on the Eastern Limb of the Bushveld Complex. A process mineralogical investigation was undertaken on ore from the Merensky Reef (MR) and the UG2 at TRP. This was conducted on a suite of geological samples (channel samples) collected from the underground workings, as well as metallurgical samples obtained from the rougher circuits at the concentrator plant during the processing of MR and UG2 ore. The geological and metallurgical samples were analysed for bulk composition and quantitative mineralogy, while the geological samples were also subjected to laboratory-scale milling and flotation tests. This study shows that, although mineralogically distinct, the MR and UG2 behave similarly in terms of metallurgical performance. This holds promise for the proposed blending of MR and UG2 ores at TRP. An evaluation of the bulk rock (ore) Pt/Pd ratio as a possible indicator of the level of hydrothermal alteration of the ore, demonstrates that this may be of use in predicting recovery plant performance.