The samples were purplish red to red in color and ranged from 1.40 to 2.17 ct.
GIA gemologists analyzed them using conventional microscopy, Fourier-transform infrared (FTIR) spectroscopy, and laser ablation–inductively coupled plasma–mass spectrometry (LA-ICP-MS).
Magnification revealed triangular growth marks, sub-parallel striations, and “heat wave”-like formations at the interface between the overgrowth and the seed. The overgrowth also contained numerous flux inclusions. Characteristics of the natural seeds included healed fractures, heat-altered crystals, and intersecting needles.
Three samples were fabricated into wafers for LA-ICP-MS analysis of their trace-element chemistry. For comparison purposes, the authors also collected chemistry data on representative synthetic ruby crystals: one flux-grown Douros and one hydrothermal example.
On the basis of microscope observation and for convenience, the authors divided the samples into two groups they dubbed “type I” and “type II.” Type I had no discernable boundary between the natural seed and the synthetic overgrowth, which was often confined to the table facet alone. Type II samples had much thicker synthetic overgrowth, with an obvious “dusty” interface between the two.
In the type I sample analyzed by LA-ICP-MS, the seed provided trace-element chemistry representative of natural corundum, while the synthetic overgrowth lacked elements such as vanadium, iron, and gallium, but had elevated levels of Cr, (up to 6,784 ppma), along with heavy elements atypical for natural corundum, including molybdenum, rhodium, and the heavy metal platinum.
The type II sample analyzed had even higher Cr levels in the synthetic overgrowth (up to 10,353 ppma), but unlike the type I example it also contained elements found in natural corundum, including magnesium, titanium, vanadium, iron, and gallium. However, manganese, nickel, zinc, and platinum were also present. This indicates different growth conditions from the type I example above. The natural seed had trace-element chemistry very similar to the type I sample described.
In both samples, the synthetic overgrowth did not correlate with the as-grown hydrothermal (Russian) and flux-grown (Douros) synthetic rubies included as a comparison.
The authors conclude with a caution not to be misled by the inclusions in the natural seed corundum, but to make a careful examination of the entire stone in order not to miss any characteristic flux inclusions within the synthetic overgrowth.