Yellow, Green, and Blue Sapphires Reportedly from Antang and Gombe, Nigeria

Yusuke Katsurada

GIA’s Tokyo laboratory borrowed 19 sapphires (figure 1) from gem dealer Ambrose & Co. in Kofu, Japan. The samples consisted of eight blue, one bluish green, eight yellow, and one greenish yellow/blue bicolor sapphires, all faceted. They were reportedly from Antang and Gombe in northern Nigeria (figure 2). Mambilla, in the southeastern part of the country, is the only well-known Nigerian sapphire source (V. Pardieu et al., “New Nigerian source of blue sapphire,” 2014, https://www.gia.edu/doc/Nigeria_Mambilla_Sapphire_US.pdf; Spring 2017 GNI, pp. 134–135), and the two new sources are not well documented.

Sapphires from both Antang and Gombe are basalt related. The mines are located in a basement complex and on Cretaceous/Tertiary sedimentary rocks near volcanic rocks. The sapphires are mined in alluvial deposits or unconsolidated sediments. According to the dealer, the blue sapphires were from Antang and the yellow sapphires and bicolor sapphire from Gombe. Brume Jeroh, the supplier in Nigeria, mentioned that Antang produces blue, green, yellow, and parti-color sapphires, while Gombe produces mainly yellow sapphires and a few blues. He also noted that stones from these sources may have been mixed together at the market in the city of Jos.

The samples’ standard gemological properties are listed in table 1. Refractive index and specific gravity values were all within the range of corundum. Representative inclusions observed in these sapphires are shown in figures 3 and 4. Quantitative analysis of trace elements was carried out with LA-ICP-MS, and the results are summarized in table 2 and figure 5.

The yellow sapphires and the greenish yellow part of the bicolor sapphire showed a typically high Fe, and their Mg concentration is greater than Ti except for one sample (N08; again, see table 2). As shown in figure 5, the high Fe (approximately 2000 ppma) and medium V (approximately 0.50–2.00 ppma) of all the yellow sapphires (N01 to N09) and the greenish yellow part of the bicolor sapphire (N10Y) showed the same trend as those of a bluish green sapphire (N11) and one blue sapphire (N12). The greenish yellow part of the bicolor sapphire (N10Y) matches other yellow sapphires in all elements, and the blue part of the bicolor (N10B) overlaps with N11 and N12. The titanium levels are significantly different between the two parts of N10. This matters, as the titanium is necessary for blue coloration in sapphire with iron. N11 and N12 show different trends from other blue sapphires (from N13 to N19) for V, Fe, and Ga. This suggests that these two samples originated from a different locality. Given that these two samples’ trace element composition is close to that of the blue part of the bicolor sapphire (N10B) and different from the rest of the blue sapphires, we considered them to be from Gombe (again, see table 1). In addition, inclusions exhibited in figure 4 in N13–N19 were not seen in N11 and N12, and these two samples had similar features to each other, as shown in figure 3. Greenish blue and blue sapphires (N11 and N12) and the blue part of the bicolor sapphire (N10B) from Gombe show high Fe (approximately 2000 ppma) and medium V (approximately 1.50–2.00 ppma). Blue sapphires from Antang (from N13 to N19) show lower Fe (approximately 800–1500 ppma) and high V (approximately 4.50 ppma). 

Although sapphires from these two new Nigerian sources are limited in quantity and color range, they are potentially high quality. The importance of the sources is not yet known, and more field data and advanced testing data such as quantitative analyses of trace elements are needed. Despite the different trends in some trace elements between the localities, as this study revealed, they still overlap with trace elements of different magmatic origins including Mambilla (data not shown). Since characteristic inclusions are not always present in basalt-related sapphires and trace elements cannot always be separated, origin determination of these sapphires remains challenging. Interpretation of trace element data such as discriminant analyses and the building of a rich database are necessary.